diff options
Diffstat (limited to 'net/netfilter')
101 files changed, 16898 insertions, 1944 deletions
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index ee898e74808..78892cf2b02 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -38,10 +38,11 @@ config NF_CONNTRACK To compile it as a module, choose M here. If unsure, say N. +if NF_CONNTRACK + config NF_CT_ACCT bool "Connection tracking flow accounting" depends on NETFILTER_ADVANCED - depends on NF_CONNTRACK help If this option is enabled, the connection tracking code will keep per-flow packet and byte counters. @@ -63,7 +64,6 @@ config NF_CT_ACCT config NF_CONNTRACK_MARK bool 'Connection mark tracking support' depends on NETFILTER_ADVANCED - depends on NF_CONNTRACK help This option enables support for connection marks, used by the `CONNMARK' target and `connmark' match. Similar to the mark value @@ -72,7 +72,7 @@ config NF_CONNTRACK_MARK config NF_CONNTRACK_SECMARK bool 'Connection tracking security mark support' - depends on NF_CONNTRACK && NETWORK_SECMARK + depends on NETWORK_SECMARK default m if NETFILTER_ADVANCED=n help This option enables security markings to be applied to @@ -85,7 +85,6 @@ config NF_CONNTRACK_SECMARK config NF_CONNTRACK_EVENTS bool "Connection tracking events" - depends on NF_CONNTRACK depends on NETFILTER_ADVANCED help If this option is enabled, the connection tracking code will @@ -96,7 +95,7 @@ config NF_CONNTRACK_EVENTS config NF_CT_PROTO_DCCP tristate 'DCCP protocol connection tracking support (EXPERIMENTAL)' - depends on EXPERIMENTAL && NF_CONNTRACK + depends on EXPERIMENTAL depends on NETFILTER_ADVANCED default IP_DCCP help @@ -107,11 +106,10 @@ config NF_CT_PROTO_DCCP config NF_CT_PROTO_GRE tristate - depends on NF_CONNTRACK config NF_CT_PROTO_SCTP tristate 'SCTP protocol connection tracking support (EXPERIMENTAL)' - depends on EXPERIMENTAL && NF_CONNTRACK + depends on EXPERIMENTAL depends on NETFILTER_ADVANCED default IP_SCTP help @@ -123,7 +121,6 @@ config NF_CT_PROTO_SCTP config NF_CT_PROTO_UDPLITE tristate 'UDP-Lite protocol connection tracking support' - depends on NF_CONNTRACK depends on NETFILTER_ADVANCED help With this option enabled, the layer 3 independent connection @@ -134,7 +131,6 @@ config NF_CT_PROTO_UDPLITE config NF_CONNTRACK_AMANDA tristate "Amanda backup protocol support" - depends on NF_CONNTRACK depends on NETFILTER_ADVANCED select TEXTSEARCH select TEXTSEARCH_KMP @@ -150,7 +146,6 @@ config NF_CONNTRACK_AMANDA config NF_CONNTRACK_FTP tristate "FTP protocol support" - depends on NF_CONNTRACK default m if NETFILTER_ADVANCED=n help Tracking FTP connections is problematic: special helpers are @@ -165,7 +160,7 @@ config NF_CONNTRACK_FTP config NF_CONNTRACK_H323 tristate "H.323 protocol support" - depends on NF_CONNTRACK && (IPV6 || IPV6=n) + depends on (IPV6 || IPV6=n) depends on NETFILTER_ADVANCED help H.323 is a VoIP signalling protocol from ITU-T. As one of the most @@ -185,7 +180,6 @@ config NF_CONNTRACK_H323 config NF_CONNTRACK_IRC tristate "IRC protocol support" - depends on NF_CONNTRACK default m if NETFILTER_ADVANCED=n help There is a commonly-used extension to IRC called @@ -201,7 +195,6 @@ config NF_CONNTRACK_IRC config NF_CONNTRACK_NETBIOS_NS tristate "NetBIOS name service protocol support" - depends on NF_CONNTRACK depends on NETFILTER_ADVANCED help NetBIOS name service requests are sent as broadcast messages from an @@ -221,7 +214,6 @@ config NF_CONNTRACK_NETBIOS_NS config NF_CONNTRACK_PPTP tristate "PPtP protocol support" - depends on NF_CONNTRACK depends on NETFILTER_ADVANCED select NF_CT_PROTO_GRE help @@ -241,7 +233,7 @@ config NF_CONNTRACK_PPTP config NF_CONNTRACK_SANE tristate "SANE protocol support (EXPERIMENTAL)" - depends on EXPERIMENTAL && NF_CONNTRACK + depends on EXPERIMENTAL depends on NETFILTER_ADVANCED help SANE is a protocol for remote access to scanners as implemented @@ -255,7 +247,6 @@ config NF_CONNTRACK_SANE config NF_CONNTRACK_SIP tristate "SIP protocol support" - depends on NF_CONNTRACK default m if NETFILTER_ADVANCED=n help SIP is an application-layer control protocol that can establish, @@ -268,7 +259,6 @@ config NF_CONNTRACK_SIP config NF_CONNTRACK_TFTP tristate "TFTP protocol support" - depends on NF_CONNTRACK depends on NETFILTER_ADVANCED help TFTP connection tracking helper, this is required depending @@ -280,13 +270,29 @@ config NF_CONNTRACK_TFTP config NF_CT_NETLINK tristate 'Connection tracking netlink interface' - depends on NF_CONNTRACK select NETFILTER_NETLINK depends on NF_NAT=n || NF_NAT default m if NETFILTER_ADVANCED=n help This option enables support for a netlink-based userspace interface +# transparent proxy support +config NETFILTER_TPROXY + tristate "Transparent proxying support (EXPERIMENTAL)" + depends on EXPERIMENTAL + depends on IP_NF_MANGLE + depends on NETFILTER_ADVANCED + help + This option enables transparent proxying support, that is, + support for handling non-locally bound IPv4 TCP and UDP sockets. + For it to work you will have to configure certain iptables rules + and use policy routing. For more information on how to set it up + see Documentation/networking/tproxy.txt. + + To compile it as a module, choose M here. If unsure, say N. + +endif # NF_CONNTRACK + config NETFILTER_XTABLES tristate "Netfilter Xtables support (required for ip_tables)" default m if NETFILTER_ADVANCED=n @@ -294,11 +300,12 @@ config NETFILTER_XTABLES This is required if you intend to use any of ip_tables, ip6_tables or arp_tables. +if NETFILTER_XTABLES + # alphabetically ordered list of targets config NETFILTER_XT_TARGET_CLASSIFY tristate '"CLASSIFY" target support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `CLASSIFY' target, which enables the user to set @@ -311,8 +318,6 @@ config NETFILTER_XT_TARGET_CLASSIFY config NETFILTER_XT_TARGET_CONNMARK tristate '"CONNMARK" target support' - depends on NETFILTER_XTABLES - depends on IP_NF_MANGLE || IP6_NF_MANGLE depends on NF_CONNTRACK depends on NETFILTER_ADVANCED select NF_CONNTRACK_MARK @@ -325,9 +330,20 @@ config NETFILTER_XT_TARGET_CONNMARK <file:Documentation/kbuild/modules.txt>. The module will be called ipt_CONNMARK.ko. If unsure, say `N'. +config NETFILTER_XT_TARGET_CONNSECMARK + tristate '"CONNSECMARK" target support' + depends on NF_CONNTRACK && NF_CONNTRACK_SECMARK + default m if NETFILTER_ADVANCED=n + help + The CONNSECMARK target copies security markings from packets + to connections, and restores security markings from connections + to packets (if the packets are not already marked). This would + normally be used in conjunction with the SECMARK target. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_TARGET_DSCP tristate '"DSCP" and "TOS" target support' - depends on NETFILTER_XTABLES depends on IP_NF_MANGLE || IP6_NF_MANGLE depends on NETFILTER_ADVANCED help @@ -344,7 +360,6 @@ config NETFILTER_XT_TARGET_DSCP config NETFILTER_XT_TARGET_MARK tristate '"MARK" target support' - depends on NETFILTER_XTABLES default m if NETFILTER_ADVANCED=n help This option adds a `MARK' target, which allows you to create rules @@ -356,21 +371,8 @@ config NETFILTER_XT_TARGET_MARK 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 - depends on NETFILTER_ADVANCED - 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_NFLOG tristate '"NFLOG" target support' - depends on NETFILTER_XTABLES default m if NETFILTER_ADVANCED=n help This option enables the NFLOG target, which allows to LOG @@ -380,9 +382,19 @@ config NETFILTER_XT_TARGET_NFLOG 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_ADVANCED + 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 NF_CONNTRACK depends on NETFILTER_ADVANCED @@ -397,7 +409,6 @@ config NETFILTER_XT_TARGET_NOTRACK config NETFILTER_XT_TARGET_RATEEST tristate '"RATEEST" target support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `RATEEST' target, which allows to measure @@ -406,9 +417,23 @@ config NETFILTER_XT_TARGET_RATEEST To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_TARGET_TPROXY + tristate '"TPROXY" target support (EXPERIMENTAL)' + depends on EXPERIMENTAL + depends on NETFILTER_TPROXY + depends on NETFILTER_XTABLES + depends on NETFILTER_ADVANCED + select NF_DEFRAG_IPV4 + help + This option adds a `TPROXY' target, which is somewhat similar to + REDIRECT. It can only be used in the mangle table and is useful + to redirect traffic to a transparent proxy. It does _not_ depend + on Netfilter connection tracking and NAT, unlike REDIRECT. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_TARGET_TRACE tristate '"TRACE" target support' - depends on NETFILTER_XTABLES depends on IP_NF_RAW || IP6_NF_RAW depends on NETFILTER_ADVANCED help @@ -421,7 +446,7 @@ config NETFILTER_XT_TARGET_TRACE config NETFILTER_XT_TARGET_SECMARK tristate '"SECMARK" target support' - depends on NETFILTER_XTABLES && NETWORK_SECMARK + depends on NETWORK_SECMARK default m if NETFILTER_ADVANCED=n help The SECMARK target allows security marking of network @@ -429,21 +454,9 @@ config NETFILTER_XT_TARGET_SECMARK To compile it as a module, choose M here. If unsure, say N. -config NETFILTER_XT_TARGET_CONNSECMARK - tristate '"CONNSECMARK" target support' - depends on NETFILTER_XTABLES && NF_CONNTRACK && NF_CONNTRACK_SECMARK - default m if NETFILTER_ADVANCED=n - help - The CONNSECMARK target copies security markings from packets - to connections, and restores security markings from connections - to packets (if the packets are not already marked). This would - normally be used in conjunction with the SECMARK target. - - To compile it as a module, choose M here. If unsure, say N. - config NETFILTER_XT_TARGET_TCPMSS tristate '"TCPMSS" target support' - depends on NETFILTER_XTABLES && (IPV6 || IPV6=n) + depends on (IPV6 || IPV6=n) default m if NETFILTER_ADVANCED=n ---help--- This option adds a `TCPMSS' target, which allows you to alter the @@ -470,7 +483,7 @@ config NETFILTER_XT_TARGET_TCPMSS config NETFILTER_XT_TARGET_TCPOPTSTRIP tristate '"TCPOPTSTRIP" target support (EXPERIMENTAL)' - depends on EXPERIMENTAL && NETFILTER_XTABLES + depends on EXPERIMENTAL depends on IP_NF_MANGLE || IP6_NF_MANGLE depends on NETFILTER_ADVANCED help @@ -479,7 +492,6 @@ config NETFILTER_XT_TARGET_TCPOPTSTRIP config NETFILTER_XT_MATCH_COMMENT tristate '"comment" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `comment' dummy-match, which allows you to put @@ -490,7 +502,6 @@ config NETFILTER_XT_MATCH_COMMENT config NETFILTER_XT_MATCH_CONNBYTES tristate '"connbytes" per-connection counter match support' - depends on NETFILTER_XTABLES depends on NF_CONNTRACK depends on NETFILTER_ADVANCED select NF_CT_ACCT @@ -503,7 +514,6 @@ config NETFILTER_XT_MATCH_CONNBYTES config NETFILTER_XT_MATCH_CONNLIMIT tristate '"connlimit" match support"' - depends on NETFILTER_XTABLES depends on NF_CONNTRACK depends on NETFILTER_ADVANCED ---help--- @@ -512,7 +522,6 @@ config NETFILTER_XT_MATCH_CONNLIMIT config NETFILTER_XT_MATCH_CONNMARK tristate '"connmark" connection mark match support' - depends on NETFILTER_XTABLES depends on NF_CONNTRACK depends on NETFILTER_ADVANCED select NF_CONNTRACK_MARK @@ -526,7 +535,6 @@ config NETFILTER_XT_MATCH_CONNMARK config NETFILTER_XT_MATCH_CONNTRACK tristate '"conntrack" connection tracking match support' - depends on NETFILTER_XTABLES depends on NF_CONNTRACK default m if NETFILTER_ADVANCED=n help @@ -540,7 +548,6 @@ config NETFILTER_XT_MATCH_CONNTRACK config NETFILTER_XT_MATCH_DCCP tristate '"dccp" protocol match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED default IP_DCCP help @@ -553,7 +560,6 @@ config NETFILTER_XT_MATCH_DCCP config NETFILTER_XT_MATCH_DSCP tristate '"dscp" and "tos" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `DSCP' match, which allows you to match against @@ -569,7 +575,6 @@ config NETFILTER_XT_MATCH_DSCP config NETFILTER_XT_MATCH_ESP tristate '"esp" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This match extension allows you to match a range of SPIs @@ -577,9 +582,23 @@ config NETFILTER_XT_MATCH_ESP To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_MATCH_HASHLIMIT + tristate '"hashlimit" match support' + depends on (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) + depends on NETFILTER_ADVANCED + help + This option adds a `hashlimit' match. + + As opposed to `limit', this match dynamically creates a hash table + of limit buckets, based on your selection of source/destination + addresses and/or ports. + + It enables you to express policies like `10kpps for any given + destination address' or `500pps from any given source address' + with a single rule. + config NETFILTER_XT_MATCH_HELPER tristate '"helper" match support' - depends on NETFILTER_XTABLES depends on NF_CONNTRACK depends on NETFILTER_ADVANCED help @@ -590,7 +609,6 @@ config NETFILTER_XT_MATCH_HELPER config NETFILTER_XT_MATCH_IPRANGE tristate '"iprange" address range match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED ---help--- This option adds a "iprange" match, which allows you to match based on @@ -601,7 +619,6 @@ config NETFILTER_XT_MATCH_IPRANGE config NETFILTER_XT_MATCH_LENGTH tristate '"length" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option allows you to match the length of a packet against a @@ -611,7 +628,6 @@ config NETFILTER_XT_MATCH_LENGTH config NETFILTER_XT_MATCH_LIMIT tristate '"limit" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help limit matching allows you to control the rate at which a rule can be @@ -622,7 +638,6 @@ config NETFILTER_XT_MATCH_LIMIT config NETFILTER_XT_MATCH_MAC tristate '"mac" address match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help MAC matching allows you to match packets based on the source @@ -632,7 +647,6 @@ config NETFILTER_XT_MATCH_MAC config NETFILTER_XT_MATCH_MARK tristate '"mark" match support' - depends on NETFILTER_XTABLES default m if NETFILTER_ADVANCED=n help Netfilter mark matching allows you to match packets based on the @@ -641,9 +655,18 @@ config NETFILTER_XT_MATCH_MARK To compile it as a module, choose M here. If unsure, say N. +config NETFILTER_XT_MATCH_MULTIPORT + tristate '"multiport" Multiple port match support' + depends on NETFILTER_ADVANCED + help + Multiport matching allows you to match TCP or UDP packets based on + a series of source or destination ports: normally a rule can only + match a single range of ports. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_MATCH_OWNER tristate '"owner" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED ---help--- Socket owner matching allows you to match locally-generated packets @@ -652,7 +675,7 @@ config NETFILTER_XT_MATCH_OWNER config NETFILTER_XT_MATCH_POLICY tristate 'IPsec "policy" match support' - depends on NETFILTER_XTABLES && XFRM + depends on XFRM default m if NETFILTER_ADVANCED=n help Policy matching allows you to match packets based on the @@ -661,20 +684,9 @@ config NETFILTER_XT_MATCH_POLICY To compile it as a module, choose M here. If unsure, say N. -config NETFILTER_XT_MATCH_MULTIPORT - tristate '"multiport" Multiple port match support' - depends on NETFILTER_XTABLES - depends on NETFILTER_ADVANCED - help - Multiport matching allows you to match TCP or UDP packets based on - a series of source or destination ports: normally a rule can only - match a single range of ports. - - 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 && BRIDGE_NETFILTER + depends on BRIDGE && BRIDGE_NETFILTER depends on NETFILTER_ADVANCED help Physdev packet matching matches against the physical bridge ports @@ -684,7 +696,6 @@ config NETFILTER_XT_MATCH_PHYSDEV config NETFILTER_XT_MATCH_PKTTYPE tristate '"pkttype" packet type match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help Packet type matching allows you to match a packet by @@ -697,7 +708,6 @@ config NETFILTER_XT_MATCH_PKTTYPE config NETFILTER_XT_MATCH_QUOTA tristate '"quota" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `quota' match, which allows to match on a @@ -708,7 +718,6 @@ config NETFILTER_XT_MATCH_QUOTA config NETFILTER_XT_MATCH_RATEEST tristate '"rateest" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED select NETFILTER_XT_TARGET_RATEEST help @@ -719,7 +728,6 @@ config NETFILTER_XT_MATCH_RATEEST config NETFILTER_XT_MATCH_REALM tristate '"realm" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED select NET_CLS_ROUTE help @@ -732,9 +740,26 @@ config NETFILTER_XT_MATCH_REALM If you want to compile it as a module, say M here and read <file:Documentation/kbuild/modules.txt>. If unsure, say `N'. +config NETFILTER_XT_MATCH_RECENT + tristate '"recent" match support' + depends on NETFILTER_ADVANCED + ---help--- + This match is used for creating one or many lists of recently + used addresses and then matching against that/those list(s). + + Short options are available by using 'iptables -m recent -h' + Official Website: <http://snowman.net/projects/ipt_recent/> + +config NETFILTER_XT_MATCH_RECENT_PROC_COMPAT + bool 'Enable obsolete /proc/net/ipt_recent' + depends on NETFILTER_XT_MATCH_RECENT && PROC_FS + ---help--- + This option enables the old /proc/net/ipt_recent interface, + which has been obsoleted by /proc/net/xt_recent. + config NETFILTER_XT_MATCH_SCTP tristate '"sctp" protocol match support (EXPERIMENTAL)' - depends on NETFILTER_XTABLES && EXPERIMENTAL + depends on EXPERIMENTAL depends on NETFILTER_ADVANCED default IP_SCTP help @@ -745,9 +770,23 @@ config NETFILTER_XT_MATCH_SCTP If you want to compile it as a module, say M here and read <file:Documentation/kbuild/modules.txt>. If unsure, say `N'. +config NETFILTER_XT_MATCH_SOCKET + tristate '"socket" match support (EXPERIMENTAL)' + depends on EXPERIMENTAL + depends on NETFILTER_TPROXY + depends on NETFILTER_XTABLES + depends on NETFILTER_ADVANCED + select NF_DEFRAG_IPV4 + help + This option adds a `socket' match, which can be used to match + packets for which a TCP or UDP socket lookup finds a valid socket. + It can be used in combination with the MARK target and policy + routing to implement full featured non-locally bound sockets. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_MATCH_STATE tristate '"state" match support' - depends on NETFILTER_XTABLES depends on NF_CONNTRACK default m if NETFILTER_ADVANCED=n help @@ -759,7 +798,6 @@ config NETFILTER_XT_MATCH_STATE config NETFILTER_XT_MATCH_STATISTIC tristate '"statistic" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `statistic' match, which allows you to match @@ -769,7 +807,6 @@ config NETFILTER_XT_MATCH_STATISTIC config NETFILTER_XT_MATCH_STRING tristate '"string" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED select TEXTSEARCH select TEXTSEARCH_KMP @@ -783,7 +820,6 @@ config NETFILTER_XT_MATCH_STRING config NETFILTER_XT_MATCH_TCPMSS tristate '"tcpmss" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED help This option adds a `tcpmss' match, which allows you to examine the @@ -794,7 +830,6 @@ config NETFILTER_XT_MATCH_TCPMSS config NETFILTER_XT_MATCH_TIME tristate '"time" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED ---help--- This option adds a "time" match, which allows you to match based on @@ -809,7 +844,6 @@ config NETFILTER_XT_MATCH_TIME config NETFILTER_XT_MATCH_U32 tristate '"u32" match support' - depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED ---help--- u32 allows you to extract quantities of up to 4 bytes from a packet, @@ -821,20 +855,8 @@ config NETFILTER_XT_MATCH_U32 Details and examples are in the kernel module source. -config NETFILTER_XT_MATCH_HASHLIMIT - tristate '"hashlimit" match support' - depends on NETFILTER_XTABLES && (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) - depends on NETFILTER_ADVANCED - help - This option adds a `hashlimit' match. - - As opposed to `limit', this match dynamically creates a hash table - of limit buckets, based on your selection of source/destination - addresses and/or ports. - - It enables you to express policies like `10kpps for any given - destination address' or `500pps from any given source address' - with a single rule. +endif # NETFILTER_XTABLES endmenu +source "net/netfilter/ipvs/Kconfig" diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 3bd2cc556ae..da3d909e053 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -34,6 +34,9 @@ obj-$(CONFIG_NF_CONNTRACK_SANE) += nf_conntrack_sane.o obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o obj-$(CONFIG_NF_CONNTRACK_TFTP) += nf_conntrack_tftp.o +# transparent proxy support +obj-$(CONFIG_NETFILTER_TPROXY) += nf_tproxy_core.o + # generic X tables obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o @@ -48,6 +51,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o obj-$(CONFIG_NETFILTER_XT_TARGET_RATEEST) += xt_RATEEST.o obj-$(CONFIG_NETFILTER_XT_TARGET_SECMARK) += xt_SECMARK.o +obj-$(CONFIG_NETFILTER_XT_TARGET_TPROXY) += xt_TPROXY.o obj-$(CONFIG_NETFILTER_XT_TARGET_TCPMSS) += xt_TCPMSS.o obj-$(CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP) += xt_TCPOPTSTRIP.o obj-$(CONFIG_NETFILTER_XT_TARGET_TRACE) += xt_TRACE.o @@ -76,10 +80,15 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_POLICY) += xt_policy.o obj-$(CONFIG_NETFILTER_XT_MATCH_QUOTA) += xt_quota.o obj-$(CONFIG_NETFILTER_XT_MATCH_RATEEST) += xt_rateest.o obj-$(CONFIG_NETFILTER_XT_MATCH_REALM) += xt_realm.o +obj-$(CONFIG_NETFILTER_XT_MATCH_RECENT) += xt_recent.o obj-$(CONFIG_NETFILTER_XT_MATCH_SCTP) += xt_sctp.o +obj-$(CONFIG_NETFILTER_XT_MATCH_SOCKET) += xt_socket.o obj-$(CONFIG_NETFILTER_XT_MATCH_STATE) += xt_state.o obj-$(CONFIG_NETFILTER_XT_MATCH_STATISTIC) += xt_statistic.o obj-$(CONFIG_NETFILTER_XT_MATCH_STRING) += xt_string.o obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o obj-$(CONFIG_NETFILTER_XT_MATCH_TIME) += xt_time.o obj-$(CONFIG_NETFILTER_XT_MATCH_U32) += xt_u32.o + +# IPVS +obj-$(CONFIG_IP_VS) += ipvs/ diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 292fa28146f..a90ac83c591 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -26,7 +26,7 @@ static DEFINE_MUTEX(afinfo_mutex); -const struct nf_afinfo *nf_afinfo[NPROTO] __read_mostly; +const struct nf_afinfo *nf_afinfo[NFPROTO_NUMPROTO] __read_mostly; EXPORT_SYMBOL(nf_afinfo); int nf_register_afinfo(const struct nf_afinfo *afinfo) @@ -51,7 +51,7 @@ void nf_unregister_afinfo(const struct nf_afinfo *afinfo) } EXPORT_SYMBOL_GPL(nf_unregister_afinfo); -struct list_head nf_hooks[NPROTO][NF_MAX_HOOKS] __read_mostly; +struct list_head nf_hooks[NFPROTO_NUMPROTO][NF_MAX_HOOKS] __read_mostly; EXPORT_SYMBOL(nf_hooks); static DEFINE_MUTEX(nf_hook_mutex); @@ -113,7 +113,7 @@ EXPORT_SYMBOL(nf_unregister_hooks); unsigned int nf_iterate(struct list_head *head, struct sk_buff *skb, - int hook, + unsigned int hook, const struct net_device *indev, const struct net_device *outdev, struct list_head **i, @@ -155,7 +155,7 @@ unsigned int nf_iterate(struct list_head *head, /* Returns 1 if okfn() needs to be executed by the caller, * -EPERM for NF_DROP, 0 otherwise. */ -int nf_hook_slow(int pf, unsigned int hook, struct sk_buff *skb, +int nf_hook_slow(u_int8_t pf, unsigned int hook, struct sk_buff *skb, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *), @@ -165,14 +165,6 @@ int nf_hook_slow(int pf, unsigned int hook, struct sk_buff *skb, unsigned int verdict; int ret = 0; -#ifdef CONFIG_NET_NS - struct net *net; - - net = indev == NULL ? dev_net(outdev) : dev_net(indev); - if (net != &init_net) - return 1; -#endif - /* We may already have this, but read-locks nest anyway */ rcu_read_lock(); @@ -264,7 +256,7 @@ EXPORT_SYMBOL(proc_net_netfilter); void __init netfilter_init(void) { int i, h; - for (i = 0; i < NPROTO; i++) { + for (i = 0; i < ARRAY_SIZE(nf_hooks); i++) { for (h = 0; h < NF_MAX_HOOKS; h++) INIT_LIST_HEAD(&nf_hooks[i][h]); } diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig new file mode 100644 index 00000000000..de6004de80b --- /dev/null +++ b/net/netfilter/ipvs/Kconfig @@ -0,0 +1,239 @@ +# +# IP Virtual Server configuration +# +menuconfig IP_VS + tristate "IP virtual server support (EXPERIMENTAL)" + depends on NETFILTER + ---help--- + IP Virtual Server support will let you build a high-performance + virtual server based on cluster of two or more real servers. This + option must be enabled for at least one of the clustered computers + that will take care of intercepting incoming connections to a + single IP address and scheduling them to real servers. + + Three request dispatching techniques are implemented, they are + virtual server via NAT, virtual server via tunneling and virtual + server via direct routing. The several scheduling algorithms can + be used to choose which server the connection is directed to, + thus load balancing can be achieved among the servers. For more + information and its administration program, please visit the + following URL: <http://www.linuxvirtualserver.org/>. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +if IP_VS + +config IP_VS_IPV6 + bool "IPv6 support for IPVS (DANGEROUS)" + depends on EXPERIMENTAL && (IPV6 = y || IP_VS = IPV6) + ---help--- + Add IPv6 support to IPVS. This is incomplete and might be dangerous. + + Say N if unsure. + +config IP_VS_DEBUG + bool "IP virtual server debugging" + ---help--- + Say Y here if you want to get additional messages useful in + debugging the IP virtual server code. You can change the debug + level in /proc/sys/net/ipv4/vs/debug_level + +config IP_VS_TAB_BITS + int "IPVS connection table size (the Nth power of 2)" + range 8 20 + default 12 + ---help--- + The IPVS connection hash table uses the chaining scheme to handle + hash collisions. Using a big IPVS connection hash table will greatly + reduce conflicts when there are hundreds of thousands of connections + in the hash table. + + Note the table size must be power of 2. The table size will be the + value of 2 to the your input number power. The number to choose is + from 8 to 20, the default number is 12, which means the table size + is 4096. Don't input the number too small, otherwise you will lose + performance on it. You can adapt the table size yourself, according + to your virtual server application. It is good to set the table size + not far less than the number of connections per second multiplying + average lasting time of connection in the table. For example, your + virtual server gets 200 connections per second, the connection lasts + for 200 seconds in average in the connection table, the table size + should be not far less than 200x200, it is good to set the table + size 32768 (2**15). + + Another note that each connection occupies 128 bytes effectively and + each hash entry uses 8 bytes, so you can estimate how much memory is + needed for your box. + +comment "IPVS transport protocol load balancing support" + +config IP_VS_PROTO_TCP + bool "TCP load balancing support" + ---help--- + This option enables support for load balancing TCP transport + protocol. Say Y if unsure. + +config IP_VS_PROTO_UDP + bool "UDP load balancing support" + ---help--- + This option enables support for load balancing UDP transport + protocol. Say Y if unsure. + +config IP_VS_PROTO_AH_ESP + bool + depends on UNDEFINED + +config IP_VS_PROTO_ESP + bool "ESP load balancing support" + select IP_VS_PROTO_AH_ESP + ---help--- + This option enables support for load balancing ESP (Encapsulation + Security Payload) transport protocol. Say Y if unsure. + +config IP_VS_PROTO_AH + bool "AH load balancing support" + select IP_VS_PROTO_AH_ESP + ---help--- + This option enables support for load balancing AH (Authentication + Header) transport protocol. Say Y if unsure. + +comment "IPVS scheduler" + +config IP_VS_RR + tristate "round-robin scheduling" + ---help--- + The robin-robin scheduling algorithm simply directs network + connections to different real servers in a round-robin manner. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_WRR + tristate "weighted round-robin scheduling" + ---help--- + The weighted robin-robin scheduling algorithm directs network + connections to different real servers based on server weights + in a round-robin manner. Servers with higher weights receive + new connections first than those with less weights, and servers + with higher weights get more connections than those with less + weights and servers with equal weights get equal connections. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_LC + tristate "least-connection scheduling" + ---help--- + The least-connection scheduling algorithm directs network + connections to the server with the least number of active + connections. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_WLC + tristate "weighted least-connection scheduling" + ---help--- + The weighted least-connection scheduling algorithm directs network + connections to the server with the least active connections + normalized by the server weight. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_LBLC + tristate "locality-based least-connection scheduling" + ---help--- + The locality-based least-connection scheduling algorithm is for + destination IP load balancing. It is usually used in cache cluster. + This algorithm usually directs packet destined for an IP address to + its server if the server is alive and under load. If the server is + overloaded (its active connection numbers is larger than its weight) + and there is a server in its half load, then allocate the weighted + least-connection server to this IP address. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_LBLCR + tristate "locality-based least-connection with replication scheduling" + ---help--- + The locality-based least-connection with replication scheduling + algorithm is also for destination IP load balancing. It is + usually used in cache cluster. It differs from the LBLC scheduling + as follows: the load balancer maintains mappings from a target + to a set of server nodes that can serve the target. Requests for + a target are assigned to the least-connection node in the target's + server set. If all the node in the server set are over loaded, + it picks up a least-connection node in the cluster and adds it + in the sever set for the target. If the server set has not been + modified for the specified time, the most loaded node is removed + from the server set, in order to avoid high degree of replication. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_DH + tristate "destination hashing scheduling" + ---help--- + The destination hashing scheduling algorithm assigns network + connections to the servers through looking up a statically assigned + hash table by their destination IP addresses. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_SH + tristate "source hashing scheduling" + ---help--- + The source hashing scheduling algorithm assigns network + connections to the servers through looking up a statically assigned + hash table by their source IP addresses. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_SED + tristate "shortest expected delay scheduling" + ---help--- + The shortest expected delay scheduling algorithm assigns network + connections to the server with the shortest expected delay. The + expected delay that the job will experience is (Ci + 1) / Ui if + sent to the ith server, in which Ci is the number of connections + on the ith server and Ui is the fixed service rate (weight) + of the ith server. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +config IP_VS_NQ + tristate "never queue scheduling" + ---help--- + The never queue scheduling algorithm adopts a two-speed model. + When there is an idle server available, the job will be sent to + the idle server, instead of waiting for a fast one. When there + is no idle server available, the job will be sent to the server + that minimize its expected delay (The Shortest Expected Delay + scheduling algorithm). + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +comment 'IPVS application helper' + +config IP_VS_FTP + tristate "FTP protocol helper" + depends on IP_VS_PROTO_TCP + ---help--- + FTP is a protocol that transfers IP address and/or port number in + the payload. In the virtual server via Network Address Translation, + the IP address and port number of real servers cannot be sent to + clients in ftp connections directly, so FTP protocol helper is + required for tracking the connection and mangling it back to that of + virtual service. + + If you want to compile it in kernel, say Y. To compile it as a + module, choose M here. If unsure, say N. + +endif # IP_VS diff --git a/net/netfilter/ipvs/Makefile b/net/netfilter/ipvs/Makefile new file mode 100644 index 00000000000..73a46fe1fe4 --- /dev/null +++ b/net/netfilter/ipvs/Makefile @@ -0,0 +1,33 @@ +# +# Makefile for the IPVS modules on top of IPv4. +# + +# IPVS transport protocol load balancing support +ip_vs_proto-objs-y := +ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_TCP) += ip_vs_proto_tcp.o +ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_UDP) += ip_vs_proto_udp.o +ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_AH_ESP) += ip_vs_proto_ah_esp.o + +ip_vs-objs := ip_vs_conn.o ip_vs_core.o ip_vs_ctl.o ip_vs_sched.o \ + ip_vs_xmit.o ip_vs_app.o ip_vs_sync.o \ + ip_vs_est.o ip_vs_proto.o \ + $(ip_vs_proto-objs-y) + + +# IPVS core +obj-$(CONFIG_IP_VS) += ip_vs.o + +# IPVS schedulers +obj-$(CONFIG_IP_VS_RR) += ip_vs_rr.o +obj-$(CONFIG_IP_VS_WRR) += ip_vs_wrr.o +obj-$(CONFIG_IP_VS_LC) += ip_vs_lc.o +obj-$(CONFIG_IP_VS_WLC) += ip_vs_wlc.o +obj-$(CONFIG_IP_VS_LBLC) += ip_vs_lblc.o +obj-$(CONFIG_IP_VS_LBLCR) += ip_vs_lblcr.o +obj-$(CONFIG_IP_VS_DH) += ip_vs_dh.o +obj-$(CONFIG_IP_VS_SH) += ip_vs_sh.o +obj-$(CONFIG_IP_VS_SED) += ip_vs_sed.o +obj-$(CONFIG_IP_VS_NQ) += ip_vs_nq.o + +# IPVS application helpers +obj-$(CONFIG_IP_VS_FTP) += ip_vs_ftp.o diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c new file mode 100644 index 00000000000..201b8ea3020 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -0,0 +1,622 @@ +/* + * ip_vs_app.c: Application module support for IPVS + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Most code here is taken from ip_masq_app.c in kernel 2.2. The difference + * is that ip_vs_app module handles the reverse direction (incoming requests + * and outgoing responses). + * + * IP_MASQ_APP application masquerading module + * + * Author: Juan Jose Ciarlante, <jjciarla@raiz.uncu.edu.ar> + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/ip.h> +#include <linux/netfilter.h> +#include <net/net_namespace.h> +#include <net/protocol.h> +#include <net/tcp.h> +#include <asm/system.h> +#include <linux/stat.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/mutex.h> + +#include <net/ip_vs.h> + +EXPORT_SYMBOL(register_ip_vs_app); +EXPORT_SYMBOL(unregister_ip_vs_app); +EXPORT_SYMBOL(register_ip_vs_app_inc); + +/* ipvs application list head */ +static LIST_HEAD(ip_vs_app_list); +static DEFINE_MUTEX(__ip_vs_app_mutex); + + +/* + * Get an ip_vs_app object + */ +static inline int ip_vs_app_get(struct ip_vs_app *app) +{ + return try_module_get(app->module); +} + + +static inline void ip_vs_app_put(struct ip_vs_app *app) +{ + module_put(app->module); +} + + +/* + * Allocate/initialize app incarnation and register it in proto apps. + */ +static int +ip_vs_app_inc_new(struct ip_vs_app *app, __u16 proto, __u16 port) +{ + struct ip_vs_protocol *pp; + struct ip_vs_app *inc; + int ret; + + if (!(pp = ip_vs_proto_get(proto))) + return -EPROTONOSUPPORT; + + if (!pp->unregister_app) + return -EOPNOTSUPP; + + inc = kmemdup(app, sizeof(*inc), GFP_KERNEL); + if (!inc) + return -ENOMEM; + INIT_LIST_HEAD(&inc->p_list); + INIT_LIST_HEAD(&inc->incs_list); + inc->app = app; + inc->port = htons(port); + atomic_set(&inc->usecnt, 0); + + if (app->timeouts) { + inc->timeout_table = + ip_vs_create_timeout_table(app->timeouts, + app->timeouts_size); + if (!inc->timeout_table) { + ret = -ENOMEM; + goto out; + } + } + + ret = pp->register_app(inc); + if (ret) + goto out; + + list_add(&inc->a_list, &app->incs_list); + IP_VS_DBG(9, "%s application %s:%u registered\n", + pp->name, inc->name, inc->port); + + return 0; + + out: + kfree(inc->timeout_table); + kfree(inc); + return ret; +} + + +/* + * Release app incarnation + */ +static void +ip_vs_app_inc_release(struct ip_vs_app *inc) +{ + struct ip_vs_protocol *pp; + + if (!(pp = ip_vs_proto_get(inc->protocol))) + return; + + if (pp->unregister_app) + pp->unregister_app(inc); + + IP_VS_DBG(9, "%s App %s:%u unregistered\n", + pp->name, inc->name, inc->port); + + list_del(&inc->a_list); + + kfree(inc->timeout_table); + kfree(inc); +} + + +/* + * Get reference to app inc (only called from softirq) + * + */ +int ip_vs_app_inc_get(struct ip_vs_app *inc) +{ + int result; + + atomic_inc(&inc->usecnt); + if (unlikely((result = ip_vs_app_get(inc->app)) != 1)) + atomic_dec(&inc->usecnt); + return result; +} + + +/* + * Put the app inc (only called from timer or net softirq) + */ +void ip_vs_app_inc_put(struct ip_vs_app *inc) +{ + ip_vs_app_put(inc->app); + atomic_dec(&inc->usecnt); +} + + +/* + * Register an application incarnation in protocol applications + */ +int +register_ip_vs_app_inc(struct ip_vs_app *app, __u16 proto, __u16 port) +{ + int result; + + mutex_lock(&__ip_vs_app_mutex); + + result = ip_vs_app_inc_new(app, proto, port); + + mutex_unlock(&__ip_vs_app_mutex); + + return result; +} + + +/* + * ip_vs_app registration routine + */ +int register_ip_vs_app(struct ip_vs_app *app) +{ + /* increase the module use count */ + ip_vs_use_count_inc(); + + mutex_lock(&__ip_vs_app_mutex); + + list_add(&app->a_list, &ip_vs_app_list); + + mutex_unlock(&__ip_vs_app_mutex); + + return 0; +} + + +/* + * ip_vs_app unregistration routine + * We are sure there are no app incarnations attached to services + */ +void unregister_ip_vs_app(struct ip_vs_app *app) +{ + struct ip_vs_app *inc, *nxt; + + mutex_lock(&__ip_vs_app_mutex); + + list_for_each_entry_safe(inc, nxt, &app->incs_list, a_list) { + ip_vs_app_inc_release(inc); + } + + list_del(&app->a_list); + + mutex_unlock(&__ip_vs_app_mutex); + + /* decrease the module use count */ + ip_vs_use_count_dec(); +} + + +/* + * Bind ip_vs_conn to its ip_vs_app (called by cp constructor) + */ +int ip_vs_bind_app(struct ip_vs_conn *cp, struct ip_vs_protocol *pp) +{ + return pp->app_conn_bind(cp); +} + + +/* + * Unbind cp from application incarnation (called by cp destructor) + */ +void ip_vs_unbind_app(struct ip_vs_conn *cp) +{ + struct ip_vs_app *inc = cp->app; + + if (!inc) + return; + + if (inc->unbind_conn) + inc->unbind_conn(inc, cp); + if (inc->done_conn) + inc->done_conn(inc, cp); + ip_vs_app_inc_put(inc); + cp->app = NULL; +} + + +/* + * Fixes th->seq based on ip_vs_seq info. + */ +static inline void vs_fix_seq(const struct ip_vs_seq *vseq, struct tcphdr *th) +{ + __u32 seq = ntohl(th->seq); + + /* + * Adjust seq with delta-offset for all packets after + * the most recent resized pkt seq and with previous_delta offset + * for all packets before most recent resized pkt seq. + */ + if (vseq->delta || vseq->previous_delta) { + if(after(seq, vseq->init_seq)) { + th->seq = htonl(seq + vseq->delta); + IP_VS_DBG(9, "vs_fix_seq(): added delta (%d) to seq\n", + vseq->delta); + } else { + th->seq = htonl(seq + vseq->previous_delta); + IP_VS_DBG(9, "vs_fix_seq(): added previous_delta " + "(%d) to seq\n", vseq->previous_delta); + } + } +} + + +/* + * Fixes th->ack_seq based on ip_vs_seq info. + */ +static inline void +vs_fix_ack_seq(const struct ip_vs_seq *vseq, struct tcphdr *th) +{ + __u32 ack_seq = ntohl(th->ack_seq); + + /* + * Adjust ack_seq with delta-offset for + * the packets AFTER most recent resized pkt has caused a shift + * for packets before most recent resized pkt, use previous_delta + */ + if (vseq->delta || vseq->previous_delta) { + /* since ack_seq is the number of octet that is expected + to receive next, so compare it with init_seq+delta */ + if(after(ack_seq, vseq->init_seq+vseq->delta)) { + th->ack_seq = htonl(ack_seq - vseq->delta); + IP_VS_DBG(9, "vs_fix_ack_seq(): subtracted delta " + "(%d) from ack_seq\n", vseq->delta); + + } else { + th->ack_seq = htonl(ack_seq - vseq->previous_delta); + IP_VS_DBG(9, "vs_fix_ack_seq(): subtracted " + "previous_delta (%d) from ack_seq\n", + vseq->previous_delta); + } + } +} + + +/* + * Updates ip_vs_seq if pkt has been resized + * Assumes already checked proto==IPPROTO_TCP and diff!=0. + */ +static inline void vs_seq_update(struct ip_vs_conn *cp, struct ip_vs_seq *vseq, + unsigned flag, __u32 seq, int diff) +{ + /* spinlock is to keep updating cp->flags atomic */ + spin_lock(&cp->lock); + if (!(cp->flags & flag) || after(seq, vseq->init_seq)) { + vseq->previous_delta = vseq->delta; + vseq->delta += diff; + vseq->init_seq = seq; + cp->flags |= flag; + } + spin_unlock(&cp->lock); +} + +static inline int app_tcp_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb, + struct ip_vs_app *app) +{ + int diff; + const unsigned int tcp_offset = ip_hdrlen(skb); + struct tcphdr *th; + __u32 seq; + + if (!skb_make_writable(skb, tcp_offset + sizeof(*th))) + return 0; + + th = (struct tcphdr *)(skb_network_header(skb) + tcp_offset); + + /* + * Remember seq number in case this pkt gets resized + */ + seq = ntohl(th->seq); + + /* + * Fix seq stuff if flagged as so. + */ + if (cp->flags & IP_VS_CONN_F_OUT_SEQ) + vs_fix_seq(&cp->out_seq, th); + if (cp->flags & IP_VS_CONN_F_IN_SEQ) + vs_fix_ack_seq(&cp->in_seq, th); + + /* + * Call private output hook function + */ + if (app->pkt_out == NULL) + return 1; + + if (!app->pkt_out(app, cp, skb, &diff)) + return 0; + + /* + * Update ip_vs seq stuff if len has changed. + */ + if (diff != 0) + vs_seq_update(cp, &cp->out_seq, + IP_VS_CONN_F_OUT_SEQ, seq, diff); + + return 1; +} + +/* + * Output pkt hook. Will call bound ip_vs_app specific function + * called by ipvs packet handler, assumes previously checked cp!=NULL + * returns false if it can't handle packet (oom) + */ +int ip_vs_app_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb) +{ + struct ip_vs_app *app; + + /* + * check if application module is bound to + * this ip_vs_conn. + */ + if ((app = cp->app) == NULL) + return 1; + + /* TCP is complicated */ + if (cp->protocol == IPPROTO_TCP) + return app_tcp_pkt_out(cp, skb, app); + + /* + * Call private output hook function + */ + if (app->pkt_out == NULL) + return 1; + + return app->pkt_out(app, cp, skb, NULL); +} + + +static inline int app_tcp_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb, + struct ip_vs_app *app) +{ + int diff; + const unsigned int tcp_offset = ip_hdrlen(skb); + struct tcphdr *th; + __u32 seq; + + if (!skb_make_writable(skb, tcp_offset + sizeof(*th))) + return 0; + + th = (struct tcphdr *)(skb_network_header(skb) + tcp_offset); + + /* + * Remember seq number in case this pkt gets resized + */ + seq = ntohl(th->seq); + + /* + * Fix seq stuff if flagged as so. + */ + if (cp->flags & IP_VS_CONN_F_IN_SEQ) + vs_fix_seq(&cp->in_seq, th); + if (cp->flags & IP_VS_CONN_F_OUT_SEQ) + vs_fix_ack_seq(&cp->out_seq, th); + + /* + * Call private input hook function + */ + if (app->pkt_in == NULL) + return 1; + + if (!app->pkt_in(app, cp, skb, &diff)) + return 0; + + /* + * Update ip_vs seq stuff if len has changed. + */ + if (diff != 0) + vs_seq_update(cp, &cp->in_seq, + IP_VS_CONN_F_IN_SEQ, seq, diff); + + return 1; +} + +/* + * Input pkt hook. Will call bound ip_vs_app specific function + * called by ipvs packet handler, assumes previously checked cp!=NULL. + * returns false if can't handle packet (oom). + */ +int ip_vs_app_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb) +{ + struct ip_vs_app *app; + + /* + * check if application module is bound to + * this ip_vs_conn. + */ + if ((app = cp->app) == NULL) + return 1; + + /* TCP is complicated */ + if (cp->protocol == IPPROTO_TCP) + return app_tcp_pkt_in(cp, skb, app); + + /* + * Call private input hook function + */ + if (app->pkt_in == NULL) + return 1; + + return app->pkt_in(app, cp, skb, NULL); +} + + +#ifdef CONFIG_PROC_FS +/* + * /proc/net/ip_vs_app entry function + */ + +static struct ip_vs_app *ip_vs_app_idx(loff_t pos) +{ + struct ip_vs_app *app, *inc; + + list_for_each_entry(app, &ip_vs_app_list, a_list) { + list_for_each_entry(inc, &app->incs_list, a_list) { + if (pos-- == 0) + return inc; + } + } + return NULL; + +} + +static void *ip_vs_app_seq_start(struct seq_file *seq, loff_t *pos) +{ + mutex_lock(&__ip_vs_app_mutex); + + return *pos ? ip_vs_app_idx(*pos - 1) : SEQ_START_TOKEN; +} + +static void *ip_vs_app_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct ip_vs_app *inc, *app; + struct list_head *e; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ip_vs_app_idx(0); + + inc = v; + app = inc->app; + + if ((e = inc->a_list.next) != &app->incs_list) + return list_entry(e, struct ip_vs_app, a_list); + + /* go on to next application */ + for (e = app->a_list.next; e != &ip_vs_app_list; e = e->next) { + app = list_entry(e, struct ip_vs_app, a_list); + list_for_each_entry(inc, &app->incs_list, a_list) { + return inc; + } + } + return NULL; +} + +static void ip_vs_app_seq_stop(struct seq_file *seq, void *v) +{ + mutex_unlock(&__ip_vs_app_mutex); +} + +static int ip_vs_app_seq_show(struct seq_file *seq, void *v) +{ + if (v == SEQ_START_TOKEN) + seq_puts(seq, "prot port usecnt name\n"); + else { + const struct ip_vs_app *inc = v; + + seq_printf(seq, "%-3s %-7u %-6d %-17s\n", + ip_vs_proto_name(inc->protocol), + ntohs(inc->port), + atomic_read(&inc->usecnt), + inc->name); + } + return 0; +} + +static const struct seq_operations ip_vs_app_seq_ops = { + .start = ip_vs_app_seq_start, + .next = ip_vs_app_seq_next, + .stop = ip_vs_app_seq_stop, + .show = ip_vs_app_seq_show, +}; + +static int ip_vs_app_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &ip_vs_app_seq_ops); +} + +static const struct file_operations ip_vs_app_fops = { + .owner = THIS_MODULE, + .open = ip_vs_app_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; +#endif + + +/* + * Replace a segment of data with a new segment + */ +int ip_vs_skb_replace(struct sk_buff *skb, gfp_t pri, + char *o_buf, int o_len, char *n_buf, int n_len) +{ + int diff; + int o_offset; + int o_left; + + EnterFunction(9); + + diff = n_len - o_len; + o_offset = o_buf - (char *)skb->data; + /* The length of left data after o_buf+o_len in the skb data */ + o_left = skb->len - (o_offset + o_len); + + if (diff <= 0) { + memmove(o_buf + n_len, o_buf + o_len, o_left); + memcpy(o_buf, n_buf, n_len); + skb_trim(skb, skb->len + diff); + } else if (diff <= skb_tailroom(skb)) { + skb_put(skb, diff); + memmove(o_buf + n_len, o_buf + o_len, o_left); + memcpy(o_buf, n_buf, n_len); + } else { + if (pskb_expand_head(skb, skb_headroom(skb), diff, pri)) + return -ENOMEM; + skb_put(skb, diff); + memmove(skb->data + o_offset + n_len, + skb->data + o_offset + o_len, o_left); + skb_copy_to_linear_data_offset(skb, o_offset, n_buf, n_len); + } + + /* must update the iph total length here */ + ip_hdr(skb)->tot_len = htons(skb->len); + + LeaveFunction(9); + return 0; +} + + +int __init ip_vs_app_init(void) +{ + /* we will replace it with proc_net_ipvs_create() soon */ + proc_net_fops_create(&init_net, "ip_vs_app", 0, &ip_vs_app_fops); + return 0; +} + + +void ip_vs_app_cleanup(void) +{ + proc_net_remove(&init_net, "ip_vs_app"); +} diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c new file mode 100644 index 00000000000..9a24332fbed --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -0,0 +1,1110 @@ +/* + * IPVS An implementation of the IP virtual server support for the + * LINUX operating system. IPVS is now implemented as a module + * over the Netfilter framework. IPVS can be used to build a + * high-performance and highly available server based on a + * cluster of servers. + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Peter Kese <peter.kese@ijs.si> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * The IPVS code for kernel 2.2 was done by Wensong Zhang and Peter Kese, + * with changes/fixes from Julian Anastasov, Lars Marowsky-Bree, Horms + * and others. Many code here is taken from IP MASQ code of kernel 2.2. + * + * Changes: + * + */ + +#include <linux/interrupt.h> +#include <linux/in.h> +#include <linux/net.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/vmalloc.h> +#include <linux/proc_fs.h> /* for proc_net_* */ +#include <linux/seq_file.h> +#include <linux/jhash.h> +#include <linux/random.h> + +#include <net/net_namespace.h> +#include <net/ip_vs.h> + + +/* + * Connection hash table: for input and output packets lookups of IPVS + */ +static struct list_head *ip_vs_conn_tab; + +/* SLAB cache for IPVS connections */ +static struct kmem_cache *ip_vs_conn_cachep __read_mostly; + +/* counter for current IPVS connections */ +static atomic_t ip_vs_conn_count = ATOMIC_INIT(0); + +/* counter for no client port connections */ +static atomic_t ip_vs_conn_no_cport_cnt = ATOMIC_INIT(0); + +/* random value for IPVS connection hash */ +static unsigned int ip_vs_conn_rnd; + +/* + * Fine locking granularity for big connection hash table + */ +#define CT_LOCKARRAY_BITS 4 +#define CT_LOCKARRAY_SIZE (1<<CT_LOCKARRAY_BITS) +#define CT_LOCKARRAY_MASK (CT_LOCKARRAY_SIZE-1) + +struct ip_vs_aligned_lock +{ + rwlock_t l; +} __attribute__((__aligned__(SMP_CACHE_BYTES))); + +/* lock array for conn table */ +static struct ip_vs_aligned_lock +__ip_vs_conntbl_lock_array[CT_LOCKARRAY_SIZE] __cacheline_aligned; + +static inline void ct_read_lock(unsigned key) +{ + read_lock(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_read_unlock(unsigned key) +{ + read_unlock(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_write_lock(unsigned key) +{ + write_lock(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_write_unlock(unsigned key) +{ + write_unlock(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_read_lock_bh(unsigned key) +{ + read_lock_bh(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_read_unlock_bh(unsigned key) +{ + read_unlock_bh(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_write_lock_bh(unsigned key) +{ + write_lock_bh(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + +static inline void ct_write_unlock_bh(unsigned key) +{ + write_unlock_bh(&__ip_vs_conntbl_lock_array[key&CT_LOCKARRAY_MASK].l); +} + + +/* + * Returns hash value for IPVS connection entry + */ +static unsigned int ip_vs_conn_hashkey(int af, unsigned proto, + const union nf_inet_addr *addr, + __be16 port) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + return jhash_3words(jhash(addr, 16, ip_vs_conn_rnd), + (__force u32)port, proto, ip_vs_conn_rnd) + & IP_VS_CONN_TAB_MASK; +#endif + return jhash_3words((__force u32)addr->ip, (__force u32)port, proto, + ip_vs_conn_rnd) + & IP_VS_CONN_TAB_MASK; +} + + +/* + * Hashes ip_vs_conn in ip_vs_conn_tab by proto,addr,port. + * returns bool success. + */ +static inline int ip_vs_conn_hash(struct ip_vs_conn *cp) +{ + unsigned hash; + int ret; + + /* Hash by protocol, client address and port */ + hash = ip_vs_conn_hashkey(cp->af, cp->protocol, &cp->caddr, cp->cport); + + ct_write_lock(hash); + + if (!(cp->flags & IP_VS_CONN_F_HASHED)) { + list_add(&cp->c_list, &ip_vs_conn_tab[hash]); + cp->flags |= IP_VS_CONN_F_HASHED; + atomic_inc(&cp->refcnt); + ret = 1; + } else { + IP_VS_ERR("ip_vs_conn_hash(): request for already hashed, " + "called from %p\n", __builtin_return_address(0)); + ret = 0; + } + + ct_write_unlock(hash); + + return ret; +} + + +/* + * UNhashes ip_vs_conn from ip_vs_conn_tab. + * returns bool success. + */ +static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp) +{ + unsigned hash; + int ret; + + /* unhash it and decrease its reference counter */ + hash = ip_vs_conn_hashkey(cp->af, cp->protocol, &cp->caddr, cp->cport); + + ct_write_lock(hash); + + if (cp->flags & IP_VS_CONN_F_HASHED) { + list_del(&cp->c_list); + cp->flags &= ~IP_VS_CONN_F_HASHED; + atomic_dec(&cp->refcnt); + ret = 1; + } else + ret = 0; + + ct_write_unlock(hash); + + return ret; +} + + +/* + * Gets ip_vs_conn associated with supplied parameters in the ip_vs_conn_tab. + * Called for pkts coming from OUTside-to-INside. + * s_addr, s_port: pkt source address (foreign host) + * d_addr, d_port: pkt dest address (load balancer) + */ +static inline struct ip_vs_conn *__ip_vs_conn_in_get +(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, + const union nf_inet_addr *d_addr, __be16 d_port) +{ + unsigned hash; + struct ip_vs_conn *cp; + + hash = ip_vs_conn_hashkey(af, protocol, s_addr, s_port); + + ct_read_lock(hash); + + list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + if (cp->af == af && + ip_vs_addr_equal(af, s_addr, &cp->caddr) && + ip_vs_addr_equal(af, d_addr, &cp->vaddr) && + s_port == cp->cport && d_port == cp->vport && + ((!s_port) ^ (!(cp->flags & IP_VS_CONN_F_NO_CPORT))) && + protocol == cp->protocol) { + /* HIT */ + atomic_inc(&cp->refcnt); + ct_read_unlock(hash); + return cp; + } + } + + ct_read_unlock(hash); + + return NULL; +} + +struct ip_vs_conn *ip_vs_conn_in_get +(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, + const union nf_inet_addr *d_addr, __be16 d_port) +{ + struct ip_vs_conn *cp; + + cp = __ip_vs_conn_in_get(af, protocol, s_addr, s_port, d_addr, d_port); + if (!cp && atomic_read(&ip_vs_conn_no_cport_cnt)) + cp = __ip_vs_conn_in_get(af, protocol, s_addr, 0, d_addr, + d_port); + + IP_VS_DBG_BUF(9, "lookup/in %s %s:%d->%s:%d %s\n", + ip_vs_proto_name(protocol), + IP_VS_DBG_ADDR(af, s_addr), ntohs(s_port), + IP_VS_DBG_ADDR(af, d_addr), ntohs(d_port), + cp ? "hit" : "not hit"); + + return cp; +} + +/* Get reference to connection template */ +struct ip_vs_conn *ip_vs_ct_in_get +(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, + const union nf_inet_addr *d_addr, __be16 d_port) +{ + unsigned hash; + struct ip_vs_conn *cp; + + hash = ip_vs_conn_hashkey(af, protocol, s_addr, s_port); + + ct_read_lock(hash); + + list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + if (cp->af == af && + ip_vs_addr_equal(af, s_addr, &cp->caddr) && + ip_vs_addr_equal(af, d_addr, &cp->vaddr) && + s_port == cp->cport && d_port == cp->vport && + cp->flags & IP_VS_CONN_F_TEMPLATE && + protocol == cp->protocol) { + /* HIT */ + atomic_inc(&cp->refcnt); + goto out; + } + } + cp = NULL; + + out: + ct_read_unlock(hash); + + IP_VS_DBG_BUF(9, "template lookup/in %s %s:%d->%s:%d %s\n", + ip_vs_proto_name(protocol), + IP_VS_DBG_ADDR(af, s_addr), ntohs(s_port), + IP_VS_DBG_ADDR(af, d_addr), ntohs(d_port), + cp ? "hit" : "not hit"); + + return cp; +} + +/* + * Gets ip_vs_conn associated with supplied parameters in the ip_vs_conn_tab. + * Called for pkts coming from inside-to-OUTside. + * s_addr, s_port: pkt source address (inside host) + * d_addr, d_port: pkt dest address (foreign host) + */ +struct ip_vs_conn *ip_vs_conn_out_get +(int af, int protocol, const union nf_inet_addr *s_addr, __be16 s_port, + const union nf_inet_addr *d_addr, __be16 d_port) +{ + unsigned hash; + struct ip_vs_conn *cp, *ret=NULL; + + /* + * Check for "full" addressed entries + */ + hash = ip_vs_conn_hashkey(af, protocol, d_addr, d_port); + + ct_read_lock(hash); + + list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + if (cp->af == af && + ip_vs_addr_equal(af, d_addr, &cp->caddr) && + ip_vs_addr_equal(af, s_addr, &cp->daddr) && + d_port == cp->cport && s_port == cp->dport && + protocol == cp->protocol) { + /* HIT */ + atomic_inc(&cp->refcnt); + ret = cp; + break; + } + } + + ct_read_unlock(hash); + + IP_VS_DBG_BUF(9, "lookup/out %s %s:%d->%s:%d %s\n", + ip_vs_proto_name(protocol), + IP_VS_DBG_ADDR(af, s_addr), ntohs(s_port), + IP_VS_DBG_ADDR(af, d_addr), ntohs(d_port), + ret ? "hit" : "not hit"); + + return ret; +} + + +/* + * Put back the conn and restart its timer with its timeout + */ +void ip_vs_conn_put(struct ip_vs_conn *cp) +{ + /* reset it expire in its timeout */ + mod_timer(&cp->timer, jiffies+cp->timeout); + + __ip_vs_conn_put(cp); +} + + +/* + * Fill a no_client_port connection with a client port number + */ +void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __be16 cport) +{ + if (ip_vs_conn_unhash(cp)) { + spin_lock(&cp->lock); + if (cp->flags & IP_VS_CONN_F_NO_CPORT) { + atomic_dec(&ip_vs_conn_no_cport_cnt); + cp->flags &= ~IP_VS_CONN_F_NO_CPORT; + cp->cport = cport; + } + spin_unlock(&cp->lock); + + /* hash on new dport */ + ip_vs_conn_hash(cp); + } +} + + +/* + * Bind a connection entry with the corresponding packet_xmit. + * Called by ip_vs_conn_new. + */ +static inline void ip_vs_bind_xmit(struct ip_vs_conn *cp) +{ + switch (IP_VS_FWD_METHOD(cp)) { + case IP_VS_CONN_F_MASQ: + cp->packet_xmit = ip_vs_nat_xmit; + break; + + case IP_VS_CONN_F_TUNNEL: + cp->packet_xmit = ip_vs_tunnel_xmit; + break; + + case IP_VS_CONN_F_DROUTE: + cp->packet_xmit = ip_vs_dr_xmit; + break; + + case IP_VS_CONN_F_LOCALNODE: + cp->packet_xmit = ip_vs_null_xmit; + break; + + case IP_VS_CONN_F_BYPASS: + cp->packet_xmit = ip_vs_bypass_xmit; + break; + } +} + +#ifdef CONFIG_IP_VS_IPV6 +static inline void ip_vs_bind_xmit_v6(struct ip_vs_conn *cp) +{ + switch (IP_VS_FWD_METHOD(cp)) { + case IP_VS_CONN_F_MASQ: + cp->packet_xmit = ip_vs_nat_xmit_v6; + break; + + case IP_VS_CONN_F_TUNNEL: + cp->packet_xmit = ip_vs_tunnel_xmit_v6; + break; + + case IP_VS_CONN_F_DROUTE: + cp->packet_xmit = ip_vs_dr_xmit_v6; + break; + + case IP_VS_CONN_F_LOCALNODE: + cp->packet_xmit = ip_vs_null_xmit; + break; + + case IP_VS_CONN_F_BYPASS: + cp->packet_xmit = ip_vs_bypass_xmit_v6; + break; + } +} +#endif + + +static inline int ip_vs_dest_totalconns(struct ip_vs_dest *dest) +{ + return atomic_read(&dest->activeconns) + + atomic_read(&dest->inactconns); +} + +/* + * Bind a connection entry with a virtual service destination + * Called just after a new connection entry is created. + */ +static inline void +ip_vs_bind_dest(struct ip_vs_conn *cp, struct ip_vs_dest *dest) +{ + /* if dest is NULL, then return directly */ + if (!dest) + return; + + /* Increase the refcnt counter of the dest */ + atomic_inc(&dest->refcnt); + + /* Bind with the destination and its corresponding transmitter */ + if ((cp->flags & IP_VS_CONN_F_SYNC) && + (!(cp->flags & IP_VS_CONN_F_TEMPLATE))) + /* if the connection is not template and is created + * by sync, preserve the activity flag. + */ + cp->flags |= atomic_read(&dest->conn_flags) & + (~IP_VS_CONN_F_INACTIVE); + else + cp->flags |= atomic_read(&dest->conn_flags); + cp->dest = dest; + + IP_VS_DBG_BUF(7, "Bind-dest %s c:%s:%d v:%s:%d " + "d:%s:%d fwd:%c s:%u conn->flags:%X conn->refcnt:%d " + "dest->refcnt:%d\n", + ip_vs_proto_name(cp->protocol), + IP_VS_DBG_ADDR(cp->af, &cp->caddr), ntohs(cp->cport), + IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport), + IP_VS_DBG_ADDR(cp->af, &cp->daddr), ntohs(cp->dport), + ip_vs_fwd_tag(cp), cp->state, + cp->flags, atomic_read(&cp->refcnt), + atomic_read(&dest->refcnt)); + + /* Update the connection counters */ + if (!(cp->flags & IP_VS_CONN_F_TEMPLATE)) { + /* It is a normal connection, so increase the inactive + connection counter because it is in TCP SYNRECV + state (inactive) or other protocol inacive state */ + if ((cp->flags & IP_VS_CONN_F_SYNC) && + (!(cp->flags & IP_VS_CONN_F_INACTIVE))) + atomic_inc(&dest->activeconns); + else + atomic_inc(&dest->inactconns); + } else { + /* It is a persistent connection/template, so increase + the peristent connection counter */ + atomic_inc(&dest->persistconns); + } + + if (dest->u_threshold != 0 && + ip_vs_dest_totalconns(dest) >= dest->u_threshold) + dest->flags |= IP_VS_DEST_F_OVERLOAD; +} + + +/* + * Check if there is a destination for the connection, if so + * bind the connection to the destination. + */ +struct ip_vs_dest *ip_vs_try_bind_dest(struct ip_vs_conn *cp) +{ + struct ip_vs_dest *dest; + + if ((cp) && (!cp->dest)) { + dest = ip_vs_find_dest(cp->af, &cp->daddr, cp->dport, + &cp->vaddr, cp->vport, + cp->protocol); + ip_vs_bind_dest(cp, dest); + return dest; + } else + return NULL; +} + + +/* + * Unbind a connection entry with its VS destination + * Called by the ip_vs_conn_expire function. + */ +static inline void ip_vs_unbind_dest(struct ip_vs_conn *cp) +{ + struct ip_vs_dest *dest = cp->dest; + + if (!dest) + return; + + IP_VS_DBG_BUF(7, "Unbind-dest %s c:%s:%d v:%s:%d " + "d:%s:%d fwd:%c s:%u conn->flags:%X conn->refcnt:%d " + "dest->refcnt:%d\n", + ip_vs_proto_name(cp->protocol), + IP_VS_DBG_ADDR(cp->af, &cp->caddr), ntohs(cp->cport), + IP_VS_DBG_ADDR(cp->af, &cp->vaddr), ntohs(cp->vport), + IP_VS_DBG_ADDR(cp->af, &cp->daddr), ntohs(cp->dport), + ip_vs_fwd_tag(cp), cp->state, + cp->flags, atomic_read(&cp->refcnt), + atomic_read(&dest->refcnt)); + + /* Update the connection counters */ + if (!(cp->flags & IP_VS_CONN_F_TEMPLATE)) { + /* It is a normal connection, so decrease the inactconns + or activeconns counter */ + if (cp->flags & IP_VS_CONN_F_INACTIVE) { + atomic_dec(&dest->inactconns); + } else { + atomic_dec(&dest->activeconns); + } + } else { + /* It is a persistent connection/template, so decrease + the peristent connection counter */ + atomic_dec(&dest->persistconns); + } + + if (dest->l_threshold != 0) { + if (ip_vs_dest_totalconns(dest) < dest->l_threshold) + dest->flags &= ~IP_VS_DEST_F_OVERLOAD; + } else if (dest->u_threshold != 0) { + if (ip_vs_dest_totalconns(dest) * 4 < dest->u_threshold * 3) + dest->flags &= ~IP_VS_DEST_F_OVERLOAD; + } else { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + dest->flags &= ~IP_VS_DEST_F_OVERLOAD; + } + + /* + * Simply decrease the refcnt of the dest, because the + * dest will be either in service's destination list + * or in the trash. + */ + atomic_dec(&dest->refcnt); +} + + +/* + * Checking if the destination of a connection template is available. + * If available, return 1, otherwise invalidate this connection + * template and return 0. + */ +int ip_vs_check_template(struct ip_vs_conn *ct) +{ + struct ip_vs_dest *dest = ct->dest; + + /* + * Checking the dest server status. + */ + if ((dest == NULL) || + !(dest->flags & IP_VS_DEST_F_AVAILABLE) || + (sysctl_ip_vs_expire_quiescent_template && + (atomic_read(&dest->weight) == 0))) { + IP_VS_DBG_BUF(9, "check_template: dest not available for " + "protocol %s s:%s:%d v:%s:%d " + "-> d:%s:%d\n", + ip_vs_proto_name(ct->protocol), + IP_VS_DBG_ADDR(ct->af, &ct->caddr), + ntohs(ct->cport), + IP_VS_DBG_ADDR(ct->af, &ct->vaddr), + ntohs(ct->vport), + IP_VS_DBG_ADDR(ct->af, &ct->daddr), + ntohs(ct->dport)); + + /* + * Invalidate the connection template + */ + if (ct->vport != htons(0xffff)) { + if (ip_vs_conn_unhash(ct)) { + ct->dport = htons(0xffff); + ct->vport = htons(0xffff); + ct->cport = 0; + ip_vs_conn_hash(ct); + } + } + + /* + * Simply decrease the refcnt of the template, + * don't restart its timer. + */ + atomic_dec(&ct->refcnt); + return 0; + } + return 1; +} + +static void ip_vs_conn_expire(unsigned long data) +{ + struct ip_vs_conn *cp = (struct ip_vs_conn *)data; + + cp->timeout = 60*HZ; + + /* + * hey, I'm using it + */ + atomic_inc(&cp->refcnt); + + /* + * do I control anybody? + */ + if (atomic_read(&cp->n_control)) + goto expire_later; + + /* + * unhash it if it is hashed in the conn table + */ + if (!ip_vs_conn_unhash(cp)) + goto expire_later; + + /* + * refcnt==1 implies I'm the only one referrer + */ + if (likely(atomic_read(&cp->refcnt) == 1)) { + /* delete the timer if it is activated by other users */ + if (timer_pending(&cp->timer)) + del_timer(&cp->timer); + + /* does anybody control me? */ + if (cp->control) + ip_vs_control_del(cp); + + if (unlikely(cp->app != NULL)) + ip_vs_unbind_app(cp); + ip_vs_unbind_dest(cp); + if (cp->flags & IP_VS_CONN_F_NO_CPORT) + atomic_dec(&ip_vs_conn_no_cport_cnt); + atomic_dec(&ip_vs_conn_count); + + kmem_cache_free(ip_vs_conn_cachep, cp); + return; + } + + /* hash it back to the table */ + ip_vs_conn_hash(cp); + + expire_later: + IP_VS_DBG(7, "delayed: conn->refcnt-1=%d conn->n_control=%d\n", + atomic_read(&cp->refcnt)-1, + atomic_read(&cp->n_control)); + + ip_vs_conn_put(cp); +} + + +void ip_vs_conn_expire_now(struct ip_vs_conn *cp) +{ + if (del_timer(&cp->timer)) + mod_timer(&cp->timer, jiffies); +} + + +/* + * Create a new connection entry and hash it into the ip_vs_conn_tab + */ +struct ip_vs_conn * +ip_vs_conn_new(int af, int proto, const union nf_inet_addr *caddr, __be16 cport, + const union nf_inet_addr *vaddr, __be16 vport, + const union nf_inet_addr *daddr, __be16 dport, unsigned flags, + struct ip_vs_dest *dest) +{ + struct ip_vs_conn *cp; + struct ip_vs_protocol *pp = ip_vs_proto_get(proto); + + cp = kmem_cache_zalloc(ip_vs_conn_cachep, GFP_ATOMIC); + if (cp == NULL) { + IP_VS_ERR_RL("ip_vs_conn_new: no memory available.\n"); + return NULL; + } + + INIT_LIST_HEAD(&cp->c_list); + setup_timer(&cp->timer, ip_vs_conn_expire, (unsigned long)cp); + cp->af = af; + cp->protocol = proto; + ip_vs_addr_copy(af, &cp->caddr, caddr); + cp->cport = cport; + ip_vs_addr_copy(af, &cp->vaddr, vaddr); + cp->vport = vport; + ip_vs_addr_copy(af, &cp->daddr, daddr); + cp->dport = dport; + cp->flags = flags; + spin_lock_init(&cp->lock); + + /* + * Set the entry is referenced by the current thread before hashing + * it in the table, so that other thread run ip_vs_random_dropentry + * but cannot drop this entry. + */ + atomic_set(&cp->refcnt, 1); + + atomic_set(&cp->n_control, 0); + atomic_set(&cp->in_pkts, 0); + + atomic_inc(&ip_vs_conn_count); + if (flags & IP_VS_CONN_F_NO_CPORT) + atomic_inc(&ip_vs_conn_no_cport_cnt); + + /* Bind the connection with a destination server */ + ip_vs_bind_dest(cp, dest); + + /* Set its state and timeout */ + cp->state = 0; + cp->timeout = 3*HZ; + + /* Bind its packet transmitter */ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + ip_vs_bind_xmit_v6(cp); + else +#endif + ip_vs_bind_xmit(cp); + + if (unlikely(pp && atomic_read(&pp->appcnt))) + ip_vs_bind_app(cp, pp); + + /* Hash it in the ip_vs_conn_tab finally */ + ip_vs_conn_hash(cp); + + return cp; +} + + +/* + * /proc/net/ip_vs_conn entries + */ +#ifdef CONFIG_PROC_FS + +static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) +{ + int idx; + struct ip_vs_conn *cp; + + for(idx = 0; idx < IP_VS_CONN_TAB_SIZE; idx++) { + ct_read_lock_bh(idx); + list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { + if (pos-- == 0) { + seq->private = &ip_vs_conn_tab[idx]; + return cp; + } + } + ct_read_unlock_bh(idx); + } + + return NULL; +} + +static void *ip_vs_conn_seq_start(struct seq_file *seq, loff_t *pos) +{ + seq->private = NULL; + return *pos ? ip_vs_conn_array(seq, *pos - 1) :SEQ_START_TOKEN; +} + +static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct ip_vs_conn *cp = v; + struct list_head *e, *l = seq->private; + int idx; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ip_vs_conn_array(seq, 0); + + /* more on same hash chain? */ + if ((e = cp->c_list.next) != l) + return list_entry(e, struct ip_vs_conn, c_list); + + idx = l - ip_vs_conn_tab; + ct_read_unlock_bh(idx); + + while (++idx < IP_VS_CONN_TAB_SIZE) { + ct_read_lock_bh(idx); + list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { + seq->private = &ip_vs_conn_tab[idx]; + return cp; + } + ct_read_unlock_bh(idx); + } + seq->private = NULL; + return NULL; +} + +static void ip_vs_conn_seq_stop(struct seq_file *seq, void *v) +{ + struct list_head *l = seq->private; + + if (l) + ct_read_unlock_bh(l - ip_vs_conn_tab); +} + +static int ip_vs_conn_seq_show(struct seq_file *seq, void *v) +{ + + if (v == SEQ_START_TOKEN) + seq_puts(seq, + "Pro FromIP FPrt ToIP TPrt DestIP DPrt State Expires\n"); + else { + const struct ip_vs_conn *cp = v; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + seq_printf(seq, + "%-3s " NIP6_FMT " %04X " NIP6_FMT + " %04X " NIP6_FMT " %04X %-11s %7lu\n", + ip_vs_proto_name(cp->protocol), + NIP6(cp->caddr.in6), ntohs(cp->cport), + NIP6(cp->vaddr.in6), ntohs(cp->vport), + NIP6(cp->daddr.in6), ntohs(cp->dport), + ip_vs_state_name(cp->protocol, cp->state), + (cp->timer.expires-jiffies)/HZ); + else +#endif + seq_printf(seq, + "%-3s %08X %04X %08X %04X" + " %08X %04X %-11s %7lu\n", + ip_vs_proto_name(cp->protocol), + ntohl(cp->caddr.ip), ntohs(cp->cport), + ntohl(cp->vaddr.ip), ntohs(cp->vport), + ntohl(cp->daddr.ip), ntohs(cp->dport), + ip_vs_state_name(cp->protocol, cp->state), + (cp->timer.expires-jiffies)/HZ); + } + return 0; +} + +static const struct seq_operations ip_vs_conn_seq_ops = { + .start = ip_vs_conn_seq_start, + .next = ip_vs_conn_seq_next, + .stop = ip_vs_conn_seq_stop, + .show = ip_vs_conn_seq_show, +}; + +static int ip_vs_conn_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &ip_vs_conn_seq_ops); +} + +static const struct file_operations ip_vs_conn_fops = { + .owner = THIS_MODULE, + .open = ip_vs_conn_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static const char *ip_vs_origin_name(unsigned flags) +{ + if (flags & IP_VS_CONN_F_SYNC) + return "SYNC"; + else + return "LOCAL"; +} + +static int ip_vs_conn_sync_seq_show(struct seq_file *seq, void *v) +{ + + if (v == SEQ_START_TOKEN) + seq_puts(seq, + "Pro FromIP FPrt ToIP TPrt DestIP DPrt State Origin Expires\n"); + else { + const struct ip_vs_conn *cp = v; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + seq_printf(seq, + "%-3s " NIP6_FMT " %04X " NIP6_FMT + " %04X " NIP6_FMT " %04X %-11s %-6s %7lu\n", + ip_vs_proto_name(cp->protocol), + NIP6(cp->caddr.in6), ntohs(cp->cport), + NIP6(cp->vaddr.in6), ntohs(cp->vport), + NIP6(cp->daddr.in6), ntohs(cp->dport), + ip_vs_state_name(cp->protocol, cp->state), + ip_vs_origin_name(cp->flags), + (cp->timer.expires-jiffies)/HZ); + else +#endif + seq_printf(seq, + "%-3s %08X %04X %08X %04X " + "%08X %04X %-11s %-6s %7lu\n", + ip_vs_proto_name(cp->protocol), + ntohl(cp->caddr.ip), ntohs(cp->cport), + ntohl(cp->vaddr.ip), ntohs(cp->vport), + ntohl(cp->daddr.ip), ntohs(cp->dport), + ip_vs_state_name(cp->protocol, cp->state), + ip_vs_origin_name(cp->flags), + (cp->timer.expires-jiffies)/HZ); + } + return 0; +} + +static const struct seq_operations ip_vs_conn_sync_seq_ops = { + .start = ip_vs_conn_seq_start, + .next = ip_vs_conn_seq_next, + .stop = ip_vs_conn_seq_stop, + .show = ip_vs_conn_sync_seq_show, +}; + +static int ip_vs_conn_sync_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &ip_vs_conn_sync_seq_ops); +} + +static const struct file_operations ip_vs_conn_sync_fops = { + .owner = THIS_MODULE, + .open = ip_vs_conn_sync_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +#endif + + +/* + * Randomly drop connection entries before running out of memory + */ +static inline int todrop_entry(struct ip_vs_conn *cp) +{ + /* + * The drop rate array needs tuning for real environments. + * Called from timer bh only => no locking + */ + static const char todrop_rate[9] = {0, 1, 2, 3, 4, 5, 6, 7, 8}; + static char todrop_counter[9] = {0}; + int i; + + /* if the conn entry hasn't lasted for 60 seconds, don't drop it. + This will leave enough time for normal connection to get + through. */ + if (time_before(cp->timeout + jiffies, cp->timer.expires + 60*HZ)) + return 0; + + /* Don't drop the entry if its number of incoming packets is not + located in [0, 8] */ + i = atomic_read(&cp->in_pkts); + if (i > 8 || i < 0) return 0; + + if (!todrop_rate[i]) return 0; + if (--todrop_counter[i] > 0) return 0; + + todrop_counter[i] = todrop_rate[i]; + return 1; +} + +/* Called from keventd and must protect itself from softirqs */ +void ip_vs_random_dropentry(void) +{ + int idx; + struct ip_vs_conn *cp; + + /* + * Randomly scan 1/32 of the whole table every second + */ + for (idx = 0; idx < (IP_VS_CONN_TAB_SIZE>>5); idx++) { + unsigned hash = net_random() & IP_VS_CONN_TAB_MASK; + + /* + * Lock is actually needed in this loop. + */ + ct_write_lock_bh(hash); + + list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + if (cp->flags & IP_VS_CONN_F_TEMPLATE) + /* connection template */ + continue; + + if (cp->protocol == IPPROTO_TCP) { + switch(cp->state) { + case IP_VS_TCP_S_SYN_RECV: + case IP_VS_TCP_S_SYNACK: + break; + + case IP_VS_TCP_S_ESTABLISHED: + if (todrop_entry(cp)) + break; + continue; + + default: + continue; + } + } else { + if (!todrop_entry(cp)) + continue; + } + + IP_VS_DBG(4, "del connection\n"); + ip_vs_conn_expire_now(cp); + if (cp->control) { + IP_VS_DBG(4, "del conn template\n"); + ip_vs_conn_expire_now(cp->control); + } + } + ct_write_unlock_bh(hash); + } +} + + +/* + * Flush all the connection entries in the ip_vs_conn_tab + */ +static void ip_vs_conn_flush(void) +{ + int idx; + struct ip_vs_conn *cp; + + flush_again: + for (idx=0; idx<IP_VS_CONN_TAB_SIZE; idx++) { + /* + * Lock is actually needed in this loop. + */ + ct_write_lock_bh(idx); + + list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { + + IP_VS_DBG(4, "del connection\n"); + ip_vs_conn_expire_now(cp); + if (cp->control) { + IP_VS_DBG(4, "del conn template\n"); + ip_vs_conn_expire_now(cp->control); + } + } + ct_write_unlock_bh(idx); + } + + /* the counter may be not NULL, because maybe some conn entries + are run by slow timer handler or unhashed but still referred */ + if (atomic_read(&ip_vs_conn_count) != 0) { + schedule(); + goto flush_again; + } +} + + +int __init ip_vs_conn_init(void) +{ + int idx; + + /* + * Allocate the connection hash table and initialize its list heads + */ + ip_vs_conn_tab = vmalloc(IP_VS_CONN_TAB_SIZE*sizeof(struct list_head)); + if (!ip_vs_conn_tab) + return -ENOMEM; + + /* Allocate ip_vs_conn slab cache */ + ip_vs_conn_cachep = kmem_cache_create("ip_vs_conn", + sizeof(struct ip_vs_conn), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!ip_vs_conn_cachep) { + vfree(ip_vs_conn_tab); + return -ENOMEM; + } + + IP_VS_INFO("Connection hash table configured " + "(size=%d, memory=%ldKbytes)\n", + IP_VS_CONN_TAB_SIZE, + (long)(IP_VS_CONN_TAB_SIZE*sizeof(struct list_head))/1024); + IP_VS_DBG(0, "Each connection entry needs %Zd bytes at least\n", + sizeof(struct ip_vs_conn)); + + for (idx = 0; idx < IP_VS_CONN_TAB_SIZE; idx++) { + INIT_LIST_HEAD(&ip_vs_conn_tab[idx]); + } + + for (idx = 0; idx < CT_LOCKARRAY_SIZE; idx++) { + rwlock_init(&__ip_vs_conntbl_lock_array[idx].l); + } + + proc_net_fops_create(&init_net, "ip_vs_conn", 0, &ip_vs_conn_fops); + proc_net_fops_create(&init_net, "ip_vs_conn_sync", 0, &ip_vs_conn_sync_fops); + + /* calculate the random value for connection hash */ + get_random_bytes(&ip_vs_conn_rnd, sizeof(ip_vs_conn_rnd)); + + return 0; +} + + +void ip_vs_conn_cleanup(void) +{ + /* flush all the connection entries first */ + ip_vs_conn_flush(); + + /* Release the empty cache */ + kmem_cache_destroy(ip_vs_conn_cachep); + proc_net_remove(&init_net, "ip_vs_conn"); + proc_net_remove(&init_net, "ip_vs_conn_sync"); + vfree(ip_vs_conn_tab); +} diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c new file mode 100644 index 00000000000..958abf3e5f8 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -0,0 +1,1542 @@ +/* + * IPVS An implementation of the IP virtual server support for the + * LINUX operating system. IPVS is now implemented as a module + * over the Netfilter framework. IPVS can be used to build a + * high-performance and highly available server based on a + * cluster of servers. + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Peter Kese <peter.kese@ijs.si> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * The IPVS code for kernel 2.2 was done by Wensong Zhang and Peter Kese, + * with changes/fixes from Julian Anastasov, Lars Marowsky-Bree, Horms + * and others. + * + * Changes: + * Paul `Rusty' Russell properly handle non-linear skbs + * Harald Welte don't use nfcache + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/ip.h> +#include <linux/tcp.h> +#include <linux/icmp.h> + +#include <net/ip.h> +#include <net/tcp.h> +#include <net/udp.h> +#include <net/icmp.h> /* for icmp_send */ +#include <net/route.h> + +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> + +#ifdef CONFIG_IP_VS_IPV6 +#include <net/ipv6.h> +#include <linux/netfilter_ipv6.h> +#endif + +#include <net/ip_vs.h> + + +EXPORT_SYMBOL(register_ip_vs_scheduler); +EXPORT_SYMBOL(unregister_ip_vs_scheduler); +EXPORT_SYMBOL(ip_vs_skb_replace); +EXPORT_SYMBOL(ip_vs_proto_name); +EXPORT_SYMBOL(ip_vs_conn_new); +EXPORT_SYMBOL(ip_vs_conn_in_get); +EXPORT_SYMBOL(ip_vs_conn_out_get); +#ifdef CONFIG_IP_VS_PROTO_TCP +EXPORT_SYMBOL(ip_vs_tcp_conn_listen); +#endif +EXPORT_SYMBOL(ip_vs_conn_put); +#ifdef CONFIG_IP_VS_DEBUG +EXPORT_SYMBOL(ip_vs_get_debug_level); +#endif + + +/* ID used in ICMP lookups */ +#define icmp_id(icmph) (((icmph)->un).echo.id) +#define icmpv6_id(icmph) (icmph->icmp6_dataun.u_echo.identifier) + +const char *ip_vs_proto_name(unsigned proto) +{ + static char buf[20]; + + switch (proto) { + case IPPROTO_IP: + return "IP"; + case IPPROTO_UDP: + return "UDP"; + case IPPROTO_TCP: + return "TCP"; + case IPPROTO_ICMP: + return "ICMP"; +#ifdef CONFIG_IP_VS_IPV6 + case IPPROTO_ICMPV6: + return "ICMPv6"; +#endif + default: + sprintf(buf, "IP_%d", proto); + return buf; + } +} + +void ip_vs_init_hash_table(struct list_head *table, int rows) +{ + while (--rows >= 0) + INIT_LIST_HEAD(&table[rows]); +} + +static inline void +ip_vs_in_stats(struct ip_vs_conn *cp, struct sk_buff *skb) +{ + struct ip_vs_dest *dest = cp->dest; + if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { + spin_lock(&dest->stats.lock); + dest->stats.ustats.inpkts++; + dest->stats.ustats.inbytes += skb->len; + spin_unlock(&dest->stats.lock); + + spin_lock(&dest->svc->stats.lock); + dest->svc->stats.ustats.inpkts++; + dest->svc->stats.ustats.inbytes += skb->len; + spin_unlock(&dest->svc->stats.lock); + + spin_lock(&ip_vs_stats.lock); + ip_vs_stats.ustats.inpkts++; + ip_vs_stats.ustats.inbytes += skb->len; + spin_unlock(&ip_vs_stats.lock); + } +} + + +static inline void +ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb) +{ + struct ip_vs_dest *dest = cp->dest; + if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { + spin_lock(&dest->stats.lock); + dest->stats.ustats.outpkts++; + dest->stats.ustats.outbytes += skb->len; + spin_unlock(&dest->stats.lock); + + spin_lock(&dest->svc->stats.lock); + dest->svc->stats.ustats.outpkts++; + dest->svc->stats.ustats.outbytes += skb->len; + spin_unlock(&dest->svc->stats.lock); + + spin_lock(&ip_vs_stats.lock); + ip_vs_stats.ustats.outpkts++; + ip_vs_stats.ustats.outbytes += skb->len; + spin_unlock(&ip_vs_stats.lock); + } +} + + +static inline void +ip_vs_conn_stats(struct ip_vs_conn *cp, struct ip_vs_service *svc) +{ + spin_lock(&cp->dest->stats.lock); + cp->dest->stats.ustats.conns++; + spin_unlock(&cp->dest->stats.lock); + + spin_lock(&svc->stats.lock); + svc->stats.ustats.conns++; + spin_unlock(&svc->stats.lock); + + spin_lock(&ip_vs_stats.lock); + ip_vs_stats.ustats.conns++; + spin_unlock(&ip_vs_stats.lock); +} + + +static inline int +ip_vs_set_state(struct ip_vs_conn *cp, int direction, + const struct sk_buff *skb, + struct ip_vs_protocol *pp) +{ + if (unlikely(!pp->state_transition)) + return 0; + return pp->state_transition(cp, direction, skb, pp); +} + + +/* + * IPVS persistent scheduling function + * It creates a connection entry according to its template if exists, + * or selects a server and creates a connection entry plus a template. + * Locking: we are svc user (svc->refcnt), so we hold all dests too + * Protocols supported: TCP, UDP + */ +static struct ip_vs_conn * +ip_vs_sched_persist(struct ip_vs_service *svc, + const struct sk_buff *skb, + __be16 ports[2]) +{ + struct ip_vs_conn *cp = NULL; + struct ip_vs_iphdr iph; + struct ip_vs_dest *dest; + struct ip_vs_conn *ct; + __be16 dport; /* destination port to forward */ + union nf_inet_addr snet; /* source network of the client, + after masking */ + + ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph); + + /* Mask saddr with the netmask to adjust template granularity */ +#ifdef CONFIG_IP_VS_IPV6 + if (svc->af == AF_INET6) + ipv6_addr_prefix(&snet.in6, &iph.saddr.in6, svc->netmask); + else +#endif + snet.ip = iph.saddr.ip & svc->netmask; + + IP_VS_DBG_BUF(6, "p-schedule: src %s:%u dest %s:%u " + "mnet %s\n", + IP_VS_DBG_ADDR(svc->af, &iph.saddr), ntohs(ports[0]), + IP_VS_DBG_ADDR(svc->af, &iph.daddr), ntohs(ports[1]), + IP_VS_DBG_ADDR(svc->af, &snet)); + + /* + * As far as we know, FTP is a very complicated network protocol, and + * it uses control connection and data connections. For active FTP, + * FTP server initialize data connection to the client, its source port + * is often 20. For passive FTP, FTP server tells the clients the port + * that it passively listens to, and the client issues the data + * connection. In the tunneling or direct routing mode, the load + * balancer is on the client-to-server half of connection, the port + * number is unknown to the load balancer. So, a conn template like + * <caddr, 0, vaddr, 0, daddr, 0> is created for persistent FTP + * service, and a template like <caddr, 0, vaddr, vport, daddr, dport> + * is created for other persistent services. + */ + if (ports[1] == svc->port) { + /* Check if a template already exists */ + if (svc->port != FTPPORT) + ct = ip_vs_ct_in_get(svc->af, iph.protocol, &snet, 0, + &iph.daddr, ports[1]); + else + ct = ip_vs_ct_in_get(svc->af, iph.protocol, &snet, 0, + &iph.daddr, 0); + + if (!ct || !ip_vs_check_template(ct)) { + /* + * No template found or the dest of the connection + * template is not available. + */ + dest = svc->scheduler->schedule(svc, skb); + if (dest == NULL) { + IP_VS_DBG(1, "p-schedule: no dest found.\n"); + return NULL; + } + + /* + * Create a template like <protocol,caddr,0, + * vaddr,vport,daddr,dport> for non-ftp service, + * and <protocol,caddr,0,vaddr,0,daddr,0> + * for ftp service. + */ + if (svc->port != FTPPORT) + ct = ip_vs_conn_new(svc->af, iph.protocol, + &snet, 0, + &iph.daddr, + ports[1], + &dest->addr, dest->port, + IP_VS_CONN_F_TEMPLATE, + dest); + else + ct = ip_vs_conn_new(svc->af, iph.protocol, + &snet, 0, + &iph.daddr, 0, + &dest->addr, 0, + IP_VS_CONN_F_TEMPLATE, + dest); + if (ct == NULL) + return NULL; + + ct->timeout = svc->timeout; + } else { + /* set destination with the found template */ + dest = ct->dest; + } + dport = dest->port; + } else { + /* + * Note: persistent fwmark-based services and persistent + * port zero service are handled here. + * fwmark template: <IPPROTO_IP,caddr,0,fwmark,0,daddr,0> + * port zero template: <protocol,caddr,0,vaddr,0,daddr,0> + */ + if (svc->fwmark) { + union nf_inet_addr fwmark = { + .all = { 0, 0, 0, htonl(svc->fwmark) } + }; + + ct = ip_vs_ct_in_get(svc->af, IPPROTO_IP, &snet, 0, + &fwmark, 0); + } else + ct = ip_vs_ct_in_get(svc->af, iph.protocol, &snet, 0, + &iph.daddr, 0); + + if (!ct || !ip_vs_check_template(ct)) { + /* + * If it is not persistent port zero, return NULL, + * otherwise create a connection template. + */ + if (svc->port) + return NULL; + + dest = svc->scheduler->schedule(svc, skb); + if (dest == NULL) { + IP_VS_DBG(1, "p-schedule: no dest found.\n"); + return NULL; + } + + /* + * Create a template according to the service + */ + if (svc->fwmark) { + union nf_inet_addr fwmark = { + .all = { 0, 0, 0, htonl(svc->fwmark) } + }; + + ct = ip_vs_conn_new(svc->af, IPPROTO_IP, + &snet, 0, + &fwmark, 0, + &dest->addr, 0, + IP_VS_CONN_F_TEMPLATE, + dest); + } else + ct = ip_vs_conn_new(svc->af, iph.protocol, + &snet, 0, + &iph.daddr, 0, + &dest->addr, 0, + IP_VS_CONN_F_TEMPLATE, + dest); + if (ct == NULL) + return NULL; + + ct->timeout = svc->timeout; + } else { + /* set destination with the found template */ + dest = ct->dest; + } + dport = ports[1]; + } + + /* + * Create a new connection according to the template + */ + cp = ip_vs_conn_new(svc->af, iph.protocol, + &iph.saddr, ports[0], + &iph.daddr, ports[1], + &dest->addr, dport, + 0, + dest); + if (cp == NULL) { + ip_vs_conn_put(ct); + return NULL; + } + + /* + * Add its control + */ + ip_vs_control_add(cp, ct); + ip_vs_conn_put(ct); + + ip_vs_conn_stats(cp, svc); + return cp; +} + + +/* + * IPVS main scheduling function + * It selects a server according to the virtual service, and + * creates a connection entry. + * Protocols supported: TCP, UDP + */ +struct ip_vs_conn * +ip_vs_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_conn *cp = NULL; + struct ip_vs_iphdr iph; + struct ip_vs_dest *dest; + __be16 _ports[2], *pptr; + + ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph); + pptr = skb_header_pointer(skb, iph.len, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + /* + * Persistent service + */ + if (svc->flags & IP_VS_SVC_F_PERSISTENT) + return ip_vs_sched_persist(svc, skb, pptr); + + /* + * Non-persistent service + */ + if (!svc->fwmark && pptr[1] != svc->port) { + if (!svc->port) + IP_VS_ERR("Schedule: port zero only supported " + "in persistent services, " + "check your ipvs configuration\n"); + return NULL; + } + + dest = svc->scheduler->schedule(svc, skb); + if (dest == NULL) { + IP_VS_DBG(1, "Schedule: no dest found.\n"); + return NULL; + } + + /* + * Create a connection entry. + */ + cp = ip_vs_conn_new(svc->af, iph.protocol, + &iph.saddr, pptr[0], + &iph.daddr, pptr[1], + &dest->addr, dest->port ? dest->port : pptr[1], + 0, + dest); + if (cp == NULL) + return NULL; + + IP_VS_DBG_BUF(6, "Schedule fwd:%c c:%s:%u v:%s:%u " + "d:%s:%u conn->flags:%X conn->refcnt:%d\n", + ip_vs_fwd_tag(cp), + IP_VS_DBG_ADDR(svc->af, &cp->caddr), ntohs(cp->cport), + IP_VS_DBG_ADDR(svc->af, &cp->vaddr), ntohs(cp->vport), + IP_VS_DBG_ADDR(svc->af, &cp->daddr), ntohs(cp->dport), + cp->flags, atomic_read(&cp->refcnt)); + + ip_vs_conn_stats(cp, svc); + return cp; +} + + +/* + * Pass or drop the packet. + * Called by ip_vs_in, when the virtual service is available but + * no destination is available for a new connection. + */ +int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, + struct ip_vs_protocol *pp) +{ + __be16 _ports[2], *pptr; + struct ip_vs_iphdr iph; + int unicast; + ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph); + + pptr = skb_header_pointer(skb, iph.len, sizeof(_ports), _ports); + if (pptr == NULL) { + ip_vs_service_put(svc); + return NF_DROP; + } + +#ifdef CONFIG_IP_VS_IPV6 + if (svc->af == AF_INET6) + unicast = ipv6_addr_type(&iph.daddr.in6) & IPV6_ADDR_UNICAST; + else +#endif + unicast = (inet_addr_type(&init_net, iph.daddr.ip) == RTN_UNICAST); + + /* if it is fwmark-based service, the cache_bypass sysctl is up + and the destination is a non-local unicast, then create + a cache_bypass connection entry */ + if (sysctl_ip_vs_cache_bypass && svc->fwmark && unicast) { + int ret, cs; + struct ip_vs_conn *cp; + union nf_inet_addr daddr = { .all = { 0, 0, 0, 0 } }; + + ip_vs_service_put(svc); + + /* create a new connection entry */ + IP_VS_DBG(6, "ip_vs_leave: create a cache_bypass entry\n"); + cp = ip_vs_conn_new(svc->af, iph.protocol, + &iph.saddr, pptr[0], + &iph.daddr, pptr[1], + &daddr, 0, + IP_VS_CONN_F_BYPASS, + NULL); + if (cp == NULL) + return NF_DROP; + + /* statistics */ + ip_vs_in_stats(cp, skb); + + /* set state */ + cs = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pp); + + /* transmit the first SYN packet */ + ret = cp->packet_xmit(skb, cp, pp); + /* do not touch skb anymore */ + + atomic_inc(&cp->in_pkts); + ip_vs_conn_put(cp); + return ret; + } + + /* + * When the virtual ftp service is presented, packets destined + * for other services on the VIP may get here (except services + * listed in the ipvs table), pass the packets, because it is + * not ipvs job to decide to drop the packets. + */ + if ((svc->port == FTPPORT) && (pptr[1] != FTPPORT)) { + ip_vs_service_put(svc); + return NF_ACCEPT; + } + + ip_vs_service_put(svc); + + /* + * Notify the client that the destination is unreachable, and + * release the socket buffer. + * Since it is in IP layer, the TCP socket is not actually + * created, the TCP RST packet cannot be sent, instead that + * ICMP_PORT_UNREACH is sent here no matter it is TCP/UDP. --WZ + */ +#ifdef CONFIG_IP_VS_IPV6 + if (svc->af == AF_INET6) + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, + skb->dev); + else +#endif + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); + + return NF_DROP; +} + + +/* + * It is hooked before NF_IP_PRI_NAT_SRC at the NF_INET_POST_ROUTING + * chain, and is used for VS/NAT. + * It detects packets for VS/NAT connections and sends the packets + * immediately. This can avoid that iptable_nat mangles the packets + * for VS/NAT. + */ +static unsigned int ip_vs_post_routing(unsigned int hooknum, + struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + if (!skb->ipvs_property) + return NF_ACCEPT; + /* The packet was sent from IPVS, exit this chain */ + return NF_STOP; +} + +__sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset) +{ + return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0)); +} + +static inline int ip_vs_gather_frags(struct sk_buff *skb, u_int32_t user) +{ + int err = ip_defrag(skb, user); + + if (!err) + ip_send_check(ip_hdr(skb)); + + return err; +} + +#ifdef CONFIG_IP_VS_IPV6 +static inline int ip_vs_gather_frags_v6(struct sk_buff *skb, u_int32_t user) +{ + /* TODO IPv6: Find out what to do here for IPv6 */ + return 0; +} +#endif + +/* + * Packet has been made sufficiently writable in caller + * - inout: 1=in->out, 0=out->in + */ +void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp, + struct ip_vs_conn *cp, int inout) +{ + struct iphdr *iph = ip_hdr(skb); + unsigned int icmp_offset = iph->ihl*4; + struct icmphdr *icmph = (struct icmphdr *)(skb_network_header(skb) + + icmp_offset); + struct iphdr *ciph = (struct iphdr *)(icmph + 1); + + if (inout) { + iph->saddr = cp->vaddr.ip; + ip_send_check(iph); + ciph->daddr = cp->vaddr.ip; + ip_send_check(ciph); + } else { + iph->daddr = cp->daddr.ip; + ip_send_check(iph); + ciph->saddr = cp->daddr.ip; + ip_send_check(ciph); + } + + /* the TCP/UDP port */ + if (IPPROTO_TCP == ciph->protocol || IPPROTO_UDP == ciph->protocol) { + __be16 *ports = (void *)ciph + ciph->ihl*4; + + if (inout) + ports[1] = cp->vport; + else + ports[0] = cp->dport; + } + + /* And finally the ICMP checksum */ + icmph->checksum = 0; + icmph->checksum = ip_vs_checksum_complete(skb, icmp_offset); + skb->ip_summed = CHECKSUM_UNNECESSARY; + + if (inout) + IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, + "Forwarding altered outgoing ICMP"); + else + IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, + "Forwarding altered incoming ICMP"); +} + +#ifdef CONFIG_IP_VS_IPV6 +void ip_vs_nat_icmp_v6(struct sk_buff *skb, struct ip_vs_protocol *pp, + struct ip_vs_conn *cp, int inout) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + unsigned int icmp_offset = sizeof(struct ipv6hdr); + struct icmp6hdr *icmph = (struct icmp6hdr *)(skb_network_header(skb) + + icmp_offset); + struct ipv6hdr *ciph = (struct ipv6hdr *)(icmph + 1); + + if (inout) { + iph->saddr = cp->vaddr.in6; + ciph->daddr = cp->vaddr.in6; + } else { + iph->daddr = cp->daddr.in6; + ciph->saddr = cp->daddr.in6; + } + + /* the TCP/UDP port */ + if (IPPROTO_TCP == ciph->nexthdr || IPPROTO_UDP == ciph->nexthdr) { + __be16 *ports = (void *)ciph + sizeof(struct ipv6hdr); + + if (inout) + ports[1] = cp->vport; + else + ports[0] = cp->dport; + } + + /* And finally the ICMP checksum */ + icmph->icmp6_cksum = 0; + /* TODO IPv6: is this correct for ICMPv6? */ + ip_vs_checksum_complete(skb, icmp_offset); + skb->ip_summed = CHECKSUM_UNNECESSARY; + + if (inout) + IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, + "Forwarding altered outgoing ICMPv6"); + else + IP_VS_DBG_PKT(11, pp, skb, (void *)ciph - (void *)iph, + "Forwarding altered incoming ICMPv6"); +} +#endif + +/* Handle relevant response ICMP messages - forward to the right + * destination host. Used for NAT and local client. + */ +static int handle_response_icmp(int af, struct sk_buff *skb, + union nf_inet_addr *snet, + __u8 protocol, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp, + unsigned int offset, unsigned int ihl) +{ + unsigned int verdict = NF_DROP; + + if (IP_VS_FWD_METHOD(cp) != 0) { + IP_VS_ERR("shouldn't reach here, because the box is on the " + "half connection in the tun/dr module.\n"); + } + + /* Ensure the checksum is correct */ + if (!skb_csum_unnecessary(skb) && ip_vs_checksum_complete(skb, ihl)) { + /* Failed checksum! */ + IP_VS_DBG_BUF(1, "Forward ICMP: failed checksum from %s!\n", + IP_VS_DBG_ADDR(af, snet)); + goto out; + } + + if (IPPROTO_TCP == protocol || IPPROTO_UDP == protocol) + offset += 2 * sizeof(__u16); + if (!skb_make_writable(skb, offset)) + goto out; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + ip_vs_nat_icmp_v6(skb, pp, cp, 1); + else +#endif + ip_vs_nat_icmp(skb, pp, cp, 1); + + /* do the statistics and put it back */ + ip_vs_out_stats(cp, skb); + + skb->ipvs_property = 1; + verdict = NF_ACCEPT; + +out: + __ip_vs_conn_put(cp); + + return verdict; +} + +/* + * Handle ICMP messages in the inside-to-outside direction (outgoing). + * Find any that might be relevant, check against existing connections. + * Currently handles error types - unreachable, quench, ttl exceeded. + */ +static int ip_vs_out_icmp(struct sk_buff *skb, int *related) +{ + struct iphdr *iph; + struct icmphdr _icmph, *ic; + struct iphdr _ciph, *cih; /* The ip header contained within the ICMP */ + struct ip_vs_iphdr ciph; + struct ip_vs_conn *cp; + struct ip_vs_protocol *pp; + unsigned int offset, ihl; + union nf_inet_addr snet; + + *related = 1; + + /* reassemble IP fragments */ + if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { + if (ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT)) + return NF_STOLEN; + } + + iph = ip_hdr(skb); + offset = ihl = iph->ihl * 4; + ic = skb_header_pointer(skb, offset, sizeof(_icmph), &_icmph); + if (ic == NULL) + return NF_DROP; + + IP_VS_DBG(12, "Outgoing ICMP (%d,%d) %u.%u.%u.%u->%u.%u.%u.%u\n", + ic->type, ntohs(icmp_id(ic)), + NIPQUAD(iph->saddr), NIPQUAD(iph->daddr)); + + /* + * Work through seeing if this is for us. + * These checks are supposed to be in an order that means easy + * things are checked first to speed up processing.... however + * this means that some packets will manage to get a long way + * down this stack and then be rejected, but that's life. + */ + if ((ic->type != ICMP_DEST_UNREACH) && + (ic->type != ICMP_SOURCE_QUENCH) && + (ic->type != ICMP_TIME_EXCEEDED)) { + *related = 0; + return NF_ACCEPT; + } + + /* Now find the contained IP header */ + offset += sizeof(_icmph); + cih = skb_header_pointer(skb, offset, sizeof(_ciph), &_ciph); + if (cih == NULL) + return NF_ACCEPT; /* The packet looks wrong, ignore */ + + pp = ip_vs_proto_get(cih->protocol); + if (!pp) + return NF_ACCEPT; + + /* Is the embedded protocol header present? */ + if (unlikely(cih->frag_off & htons(IP_OFFSET) && + pp->dont_defrag)) + return NF_ACCEPT; + + IP_VS_DBG_PKT(11, pp, skb, offset, "Checking outgoing ICMP for"); + + offset += cih->ihl * 4; + + ip_vs_fill_iphdr(AF_INET, cih, &ciph); + /* The embedded headers contain source and dest in reverse order */ + cp = pp->conn_out_get(AF_INET, skb, pp, &ciph, offset, 1); + if (!cp) + return NF_ACCEPT; + + snet.ip = iph->saddr; + return handle_response_icmp(AF_INET, skb, &snet, cih->protocol, cp, + pp, offset, ihl); +} + +#ifdef CONFIG_IP_VS_IPV6 +static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related) +{ + struct ipv6hdr *iph; + struct icmp6hdr _icmph, *ic; + struct ipv6hdr _ciph, *cih; /* The ip header contained + within the ICMP */ + struct ip_vs_iphdr ciph; + struct ip_vs_conn *cp; + struct ip_vs_protocol *pp; + unsigned int offset; + union nf_inet_addr snet; + + *related = 1; + + /* reassemble IP fragments */ + if (ipv6_hdr(skb)->nexthdr == IPPROTO_FRAGMENT) { + if (ip_vs_gather_frags_v6(skb, IP_DEFRAG_VS_OUT)) + return NF_STOLEN; + } + + iph = ipv6_hdr(skb); + offset = sizeof(struct ipv6hdr); + ic = skb_header_pointer(skb, offset, sizeof(_icmph), &_icmph); + if (ic == NULL) + return NF_DROP; + + IP_VS_DBG(12, "Outgoing ICMPv6 (%d,%d) " NIP6_FMT "->" NIP6_FMT "\n", + ic->icmp6_type, ntohs(icmpv6_id(ic)), + NIP6(iph->saddr), NIP6(iph->daddr)); + + /* + * Work through seeing if this is for us. + * These checks are supposed to be in an order that means easy + * things are checked first to speed up processing.... however + * this means that some packets will manage to get a long way + * down this stack and then be rejected, but that's life. + */ + if ((ic->icmp6_type != ICMPV6_DEST_UNREACH) && + (ic->icmp6_type != ICMPV6_PKT_TOOBIG) && + (ic->icmp6_type != ICMPV6_TIME_EXCEED)) { + *related = 0; + return NF_ACCEPT; + } + + /* Now find the contained IP header */ + offset += sizeof(_icmph); + cih = skb_header_pointer(skb, offset, sizeof(_ciph), &_ciph); + if (cih == NULL) + return NF_ACCEPT; /* The packet looks wrong, ignore */ + + pp = ip_vs_proto_get(cih->nexthdr); + if (!pp) + return NF_ACCEPT; + + /* Is the embedded protocol header present? */ + /* TODO: we don't support fragmentation at the moment anyways */ + if (unlikely(cih->nexthdr == IPPROTO_FRAGMENT && pp->dont_defrag)) + return NF_ACCEPT; + + IP_VS_DBG_PKT(11, pp, skb, offset, "Checking outgoing ICMPv6 for"); + + offset += sizeof(struct ipv6hdr); + + ip_vs_fill_iphdr(AF_INET6, cih, &ciph); + /* The embedded headers contain source and dest in reverse order */ + cp = pp->conn_out_get(AF_INET6, skb, pp, &ciph, offset, 1); + if (!cp) + return NF_ACCEPT; + + ipv6_addr_copy(&snet.in6, &iph->saddr); + return handle_response_icmp(AF_INET6, skb, &snet, cih->nexthdr, cp, + pp, offset, sizeof(struct ipv6hdr)); +} +#endif + +static inline int is_tcp_reset(const struct sk_buff *skb, int nh_len) +{ + struct tcphdr _tcph, *th; + + th = skb_header_pointer(skb, nh_len, sizeof(_tcph), &_tcph); + if (th == NULL) + return 0; + return th->rst; +} + +/* Handle response packets: rewrite addresses and send away... + * Used for NAT and local client. + */ +static unsigned int +handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, + struct ip_vs_conn *cp, int ihl) +{ + IP_VS_DBG_PKT(11, pp, skb, 0, "Outgoing packet"); + + if (!skb_make_writable(skb, ihl)) + goto drop; + + /* mangle the packet */ + if (pp->snat_handler && !pp->snat_handler(skb, pp, cp)) + goto drop; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + ipv6_hdr(skb)->saddr = cp->vaddr.in6; + else +#endif + { + ip_hdr(skb)->saddr = cp->vaddr.ip; + ip_send_check(ip_hdr(skb)); + } + + /* For policy routing, packets originating from this + * machine itself may be routed differently to packets + * passing through. We want this packet to be routed as + * if it came from this machine itself. So re-compute + * the routing information. + */ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (ip6_route_me_harder(skb) != 0) + goto drop; + } else +#endif + if (ip_route_me_harder(skb, RTN_LOCAL) != 0) + goto drop; + + IP_VS_DBG_PKT(10, pp, skb, 0, "After SNAT"); + + ip_vs_out_stats(cp, skb); + ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pp); + ip_vs_conn_put(cp); + + skb->ipvs_property = 1; + + LeaveFunction(11); + return NF_ACCEPT; + +drop: + ip_vs_conn_put(cp); + kfree_skb(skb); + return NF_STOLEN; +} + +/* + * It is hooked at the NF_INET_FORWARD chain, used only for VS/NAT. + * Check if outgoing packet belongs to the established ip_vs_conn. + */ +static unsigned int +ip_vs_out(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct ip_vs_iphdr iph; + struct ip_vs_protocol *pp; + struct ip_vs_conn *cp; + int af; + + EnterFunction(11); + + af = (skb->protocol == htons(ETH_P_IP)) ? AF_INET : AF_INET6; + + if (skb->ipvs_property) + return NF_ACCEPT; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (unlikely(iph.protocol == IPPROTO_ICMPV6)) { + int related, verdict = ip_vs_out_icmp_v6(skb, &related); + + if (related) + return verdict; + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + } + } else +#endif + if (unlikely(iph.protocol == IPPROTO_ICMP)) { + int related, verdict = ip_vs_out_icmp(skb, &related); + + if (related) + return verdict; + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + } + + pp = ip_vs_proto_get(iph.protocol); + if (unlikely(!pp)) + return NF_ACCEPT; + + /* reassemble IP fragments */ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (unlikely(iph.protocol == IPPROTO_ICMPV6)) { + int related, verdict = ip_vs_out_icmp_v6(skb, &related); + + if (related) + return verdict; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + } + } else +#endif + if (unlikely(ip_hdr(skb)->frag_off & htons(IP_MF|IP_OFFSET) && + !pp->dont_defrag)) { + if (ip_vs_gather_frags(skb, IP_DEFRAG_VS_OUT)) + return NF_STOLEN; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + } + + /* + * Check if the packet belongs to an existing entry + */ + cp = pp->conn_out_get(af, skb, pp, &iph, iph.len, 0); + + if (unlikely(!cp)) { + if (sysctl_ip_vs_nat_icmp_send && + (pp->protocol == IPPROTO_TCP || + pp->protocol == IPPROTO_UDP)) { + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, iph.len, + sizeof(_ports), _ports); + if (pptr == NULL) + return NF_ACCEPT; /* Not for me */ + if (ip_vs_lookup_real_service(af, iph.protocol, + &iph.saddr, + pptr[0])) { + /* + * Notify the real server: there is no + * existing entry if it is not RST + * packet or not TCP packet. + */ + if (iph.protocol != IPPROTO_TCP + || !is_tcp_reset(skb, iph.len)) { +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + icmpv6_send(skb, + ICMPV6_DEST_UNREACH, + ICMPV6_PORT_UNREACH, + 0, skb->dev); + else +#endif + icmp_send(skb, + ICMP_DEST_UNREACH, + ICMP_PORT_UNREACH, 0); + return NF_DROP; + } + } + } + IP_VS_DBG_PKT(12, pp, skb, 0, + "packet continues traversal as normal"); + return NF_ACCEPT; + } + + return handle_response(af, skb, pp, cp, iph.len); +} + + +/* + * Handle ICMP messages in the outside-to-inside direction (incoming). + * Find any that might be relevant, check against existing connections, + * forward to the right destination host if relevant. + * Currently handles error types - unreachable, quench, ttl exceeded. + */ +static int +ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) +{ + struct iphdr *iph; + struct icmphdr _icmph, *ic; + struct iphdr _ciph, *cih; /* The ip header contained within the ICMP */ + struct ip_vs_iphdr ciph; + struct ip_vs_conn *cp; + struct ip_vs_protocol *pp; + unsigned int offset, ihl, verdict; + union nf_inet_addr snet; + + *related = 1; + + /* reassemble IP fragments */ + if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { + if (ip_vs_gather_frags(skb, hooknum == NF_INET_LOCAL_IN ? + IP_DEFRAG_VS_IN : IP_DEFRAG_VS_FWD)) + return NF_STOLEN; + } + + iph = ip_hdr(skb); + offset = ihl = iph->ihl * 4; + ic = skb_header_pointer(skb, offset, sizeof(_icmph), &_icmph); + if (ic == NULL) + return NF_DROP; + + IP_VS_DBG(12, "Incoming ICMP (%d,%d) %u.%u.%u.%u->%u.%u.%u.%u\n", + ic->type, ntohs(icmp_id(ic)), + NIPQUAD(iph->saddr), NIPQUAD(iph->daddr)); + + /* + * Work through seeing if this is for us. + * These checks are supposed to be in an order that means easy + * things are checked first to speed up processing.... however + * this means that some packets will manage to get a long way + * down this stack and then be rejected, but that's life. + */ + if ((ic->type != ICMP_DEST_UNREACH) && + (ic->type != ICMP_SOURCE_QUENCH) && + (ic->type != ICMP_TIME_EXCEEDED)) { + *related = 0; + return NF_ACCEPT; + } + + /* Now find the contained IP header */ + offset += sizeof(_icmph); + cih = skb_header_pointer(skb, offset, sizeof(_ciph), &_ciph); + if (cih == NULL) + return NF_ACCEPT; /* The packet looks wrong, ignore */ + + pp = ip_vs_proto_get(cih->protocol); + if (!pp) + return NF_ACCEPT; + + /* Is the embedded protocol header present? */ + if (unlikely(cih->frag_off & htons(IP_OFFSET) && + pp->dont_defrag)) + return NF_ACCEPT; + + IP_VS_DBG_PKT(11, pp, skb, offset, "Checking incoming ICMP for"); + + offset += cih->ihl * 4; + + ip_vs_fill_iphdr(AF_INET, cih, &ciph); + /* The embedded headers contain source and dest in reverse order */ + cp = pp->conn_in_get(AF_INET, skb, pp, &ciph, offset, 1); + if (!cp) { + /* The packet could also belong to a local client */ + cp = pp->conn_out_get(AF_INET, skb, pp, &ciph, offset, 1); + if (cp) { + snet.ip = iph->saddr; + return handle_response_icmp(AF_INET, skb, &snet, + cih->protocol, cp, pp, + offset, ihl); + } + return NF_ACCEPT; + } + + verdict = NF_DROP; + + /* Ensure the checksum is correct */ + if (!skb_csum_unnecessary(skb) && ip_vs_checksum_complete(skb, ihl)) { + /* Failed checksum! */ + IP_VS_DBG(1, "Incoming ICMP: failed checksum from %d.%d.%d.%d!\n", + NIPQUAD(iph->saddr)); + goto out; + } + + /* do the statistics and put it back */ + ip_vs_in_stats(cp, skb); + if (IPPROTO_TCP == cih->protocol || IPPROTO_UDP == cih->protocol) + offset += 2 * sizeof(__u16); + verdict = ip_vs_icmp_xmit(skb, cp, pp, offset); + /* do not touch skb anymore */ + + out: + __ip_vs_conn_put(cp); + + return verdict; +} + +#ifdef CONFIG_IP_VS_IPV6 +static int +ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) +{ + struct ipv6hdr *iph; + struct icmp6hdr _icmph, *ic; + struct ipv6hdr _ciph, *cih; /* The ip header contained + within the ICMP */ + struct ip_vs_iphdr ciph; + struct ip_vs_conn *cp; + struct ip_vs_protocol *pp; + unsigned int offset, verdict; + union nf_inet_addr snet; + + *related = 1; + + /* reassemble IP fragments */ + if (ipv6_hdr(skb)->nexthdr == IPPROTO_FRAGMENT) { + if (ip_vs_gather_frags_v6(skb, hooknum == NF_INET_LOCAL_IN ? + IP_DEFRAG_VS_IN : + IP_DEFRAG_VS_FWD)) + return NF_STOLEN; + } + + iph = ipv6_hdr(skb); + offset = sizeof(struct ipv6hdr); + ic = skb_header_pointer(skb, offset, sizeof(_icmph), &_icmph); + if (ic == NULL) + return NF_DROP; + + IP_VS_DBG(12, "Incoming ICMPv6 (%d,%d) " NIP6_FMT "->" NIP6_FMT "\n", + ic->icmp6_type, ntohs(icmpv6_id(ic)), + NIP6(iph->saddr), NIP6(iph->daddr)); + + /* + * Work through seeing if this is for us. + * These checks are supposed to be in an order that means easy + * things are checked first to speed up processing.... however + * this means that some packets will manage to get a long way + * down this stack and then be rejected, but that's life. + */ + if ((ic->icmp6_type != ICMPV6_DEST_UNREACH) && + (ic->icmp6_type != ICMPV6_PKT_TOOBIG) && + (ic->icmp6_type != ICMPV6_TIME_EXCEED)) { + *related = 0; + return NF_ACCEPT; + } + + /* Now find the contained IP header */ + offset += sizeof(_icmph); + cih = skb_header_pointer(skb, offset, sizeof(_ciph), &_ciph); + if (cih == NULL) + return NF_ACCEPT; /* The packet looks wrong, ignore */ + + pp = ip_vs_proto_get(cih->nexthdr); + if (!pp) + return NF_ACCEPT; + + /* Is the embedded protocol header present? */ + /* TODO: we don't support fragmentation at the moment anyways */ + if (unlikely(cih->nexthdr == IPPROTO_FRAGMENT && pp->dont_defrag)) + return NF_ACCEPT; + + IP_VS_DBG_PKT(11, pp, skb, offset, "Checking incoming ICMPv6 for"); + + offset += sizeof(struct ipv6hdr); + + ip_vs_fill_iphdr(AF_INET6, cih, &ciph); + /* The embedded headers contain source and dest in reverse order */ + cp = pp->conn_in_get(AF_INET6, skb, pp, &ciph, offset, 1); + if (!cp) { + /* The packet could also belong to a local client */ + cp = pp->conn_out_get(AF_INET6, skb, pp, &ciph, offset, 1); + if (cp) { + ipv6_addr_copy(&snet.in6, &iph->saddr); + return handle_response_icmp(AF_INET6, skb, &snet, + cih->nexthdr, + cp, pp, offset, + sizeof(struct ipv6hdr)); + } + return NF_ACCEPT; + } + + verdict = NF_DROP; + + /* do the statistics and put it back */ + ip_vs_in_stats(cp, skb); + if (IPPROTO_TCP == cih->nexthdr || IPPROTO_UDP == cih->nexthdr) + offset += 2 * sizeof(__u16); + verdict = ip_vs_icmp_xmit_v6(skb, cp, pp, offset); + /* do not touch skb anymore */ + + __ip_vs_conn_put(cp); + + return verdict; +} +#endif + + +/* + * Check if it's for virtual services, look it up, + * and send it on its way... + */ +static unsigned int +ip_vs_in(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct ip_vs_iphdr iph; + struct ip_vs_protocol *pp; + struct ip_vs_conn *cp; + int ret, restart, af; + + af = (skb->protocol == htons(ETH_P_IP)) ? AF_INET : AF_INET6; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + + /* + * Big tappo: only PACKET_HOST, including loopback for local client + * Don't handle local packets on IPv6 for now + */ + if (unlikely(skb->pkt_type != PACKET_HOST)) { + IP_VS_DBG_BUF(12, "packet type=%d proto=%d daddr=%s ignored\n", + skb->pkt_type, + iph.protocol, + IP_VS_DBG_ADDR(af, &iph.daddr)); + return NF_ACCEPT; + } + + if (unlikely(iph.protocol == IPPROTO_ICMP)) { + int related, verdict = ip_vs_in_icmp(skb, &related, hooknum); + + if (related) + return verdict; + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + } + + /* Protocol supported? */ + pp = ip_vs_proto_get(iph.protocol); + if (unlikely(!pp)) + return NF_ACCEPT; + + /* + * Check if the packet belongs to an existing connection entry + */ + cp = pp->conn_in_get(af, skb, pp, &iph, iph.len, 0); + + if (unlikely(!cp)) { + int v; + + /* For local client packets, it could be a response */ + cp = pp->conn_out_get(af, skb, pp, &iph, iph.len, 0); + if (cp) + return handle_response(af, skb, pp, cp, iph.len); + + if (!pp->conn_schedule(af, skb, pp, &v, &cp)) + return v; + } + + if (unlikely(!cp)) { + /* sorry, all this trouble for a no-hit :) */ + IP_VS_DBG_PKT(12, pp, skb, 0, + "packet continues traversal as normal"); + return NF_ACCEPT; + } + + IP_VS_DBG_PKT(11, pp, skb, 0, "Incoming packet"); + + /* Check the server status */ + if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) { + /* the destination server is not available */ + + if (sysctl_ip_vs_expire_nodest_conn) { + /* try to expire the connection immediately */ + ip_vs_conn_expire_now(cp); + } + /* don't restart its timer, and silently + drop the packet. */ + __ip_vs_conn_put(cp); + return NF_DROP; + } + + ip_vs_in_stats(cp, skb); + restart = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pp); + if (cp->packet_xmit) + ret = cp->packet_xmit(skb, cp, pp); + /* do not touch skb anymore */ + else { + IP_VS_DBG_RL("warning: packet_xmit is null"); + ret = NF_ACCEPT; + } + + /* Increase its packet counter and check if it is needed + * to be synchronized + * + * Sync connection if it is about to close to + * encorage the standby servers to update the connections timeout + */ + atomic_inc(&cp->in_pkts); + if (af == AF_INET && + (ip_vs_sync_state & IP_VS_STATE_MASTER) && + (((cp->protocol != IPPROTO_TCP || + cp->state == IP_VS_TCP_S_ESTABLISHED) && + (atomic_read(&cp->in_pkts) % sysctl_ip_vs_sync_threshold[1] + == sysctl_ip_vs_sync_threshold[0])) || + ((cp->protocol == IPPROTO_TCP) && (cp->old_state != cp->state) && + ((cp->state == IP_VS_TCP_S_FIN_WAIT) || + (cp->state == IP_VS_TCP_S_CLOSE_WAIT) || + (cp->state == IP_VS_TCP_S_TIME_WAIT))))) + ip_vs_sync_conn(cp); + cp->old_state = cp->state; + + ip_vs_conn_put(cp); + return ret; +} + + +/* + * It is hooked at the NF_INET_FORWARD chain, in order to catch ICMP + * related packets destined for 0.0.0.0/0. + * When fwmark-based virtual service is used, such as transparent + * cache cluster, TCP packets can be marked and routed to ip_vs_in, + * but ICMP destined for 0.0.0.0/0 cannot not be easily marked and + * sent to ip_vs_in_icmp. So, catch them at the NF_INET_FORWARD chain + * and send them to ip_vs_in_icmp. + */ +static unsigned int +ip_vs_forward_icmp(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + int r; + + if (ip_hdr(skb)->protocol != IPPROTO_ICMP) + return NF_ACCEPT; + + return ip_vs_in_icmp(skb, &r, hooknum); +} + +#ifdef CONFIG_IP_VS_IPV6 +static unsigned int +ip_vs_forward_icmp_v6(unsigned int hooknum, struct sk_buff *skb, + const struct net_device *in, const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + int r; + + if (ipv6_hdr(skb)->nexthdr != IPPROTO_ICMPV6) + return NF_ACCEPT; + + return ip_vs_in_icmp_v6(skb, &r, hooknum); +} +#endif + + +static struct nf_hook_ops ip_vs_ops[] __read_mostly = { + /* After packet filtering, forward packet through VS/DR, VS/TUN, + * or VS/NAT(change destination), so that filtering rules can be + * applied to IPVS. */ + { + .hook = ip_vs_in, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_LOCAL_IN, + .priority = 100, + }, + /* After packet filtering, change source only for VS/NAT */ + { + .hook = ip_vs_out, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_FORWARD, + .priority = 100, + }, + /* After packet filtering (but before ip_vs_out_icmp), catch icmp + * destined for 0.0.0.0/0, which is for incoming IPVS connections */ + { + .hook = ip_vs_forward_icmp, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_FORWARD, + .priority = 99, + }, + /* Before the netfilter connection tracking, exit from POST_ROUTING */ + { + .hook = ip_vs_post_routing, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_INET_POST_ROUTING, + .priority = NF_IP_PRI_NAT_SRC-1, + }, +#ifdef CONFIG_IP_VS_IPV6 + /* After packet filtering, forward packet through VS/DR, VS/TUN, + * or VS/NAT(change destination), so that filtering rules can be + * applied to IPVS. */ + { + .hook = ip_vs_in, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_INET_LOCAL_IN, + .priority = 100, + }, + /* After packet filtering, change source only for VS/NAT */ + { + .hook = ip_vs_out, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_INET_FORWARD, + .priority = 100, + }, + /* After packet filtering (but before ip_vs_out_icmp), catch icmp + * destined for 0.0.0.0/0, which is for incoming IPVS connections */ + { + .hook = ip_vs_forward_icmp_v6, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_INET_FORWARD, + .priority = 99, + }, + /* Before the netfilter connection tracking, exit from POST_ROUTING */ + { + .hook = ip_vs_post_routing, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_INET_POST_ROUTING, + .priority = NF_IP6_PRI_NAT_SRC-1, + }, +#endif +}; + + +/* + * Initialize IP Virtual Server + */ +static int __init ip_vs_init(void) +{ + int ret; + + ip_vs_estimator_init(); + + ret = ip_vs_control_init(); + if (ret < 0) { + IP_VS_ERR("can't setup control.\n"); + goto cleanup_estimator; + } + + ip_vs_protocol_init(); + + ret = ip_vs_app_init(); + if (ret < 0) { + IP_VS_ERR("can't setup application helper.\n"); + goto cleanup_protocol; + } + + ret = ip_vs_conn_init(); + if (ret < 0) { + IP_VS_ERR("can't setup connection table.\n"); + goto cleanup_app; + } + + ret = nf_register_hooks(ip_vs_ops, ARRAY_SIZE(ip_vs_ops)); + if (ret < 0) { + IP_VS_ERR("can't register hooks.\n"); + goto cleanup_conn; + } + + IP_VS_INFO("ipvs loaded.\n"); + return ret; + + cleanup_conn: + ip_vs_conn_cleanup(); + cleanup_app: + ip_vs_app_cleanup(); + cleanup_protocol: + ip_vs_protocol_cleanup(); + ip_vs_control_cleanup(); + cleanup_estimator: + ip_vs_estimator_cleanup(); + return ret; +} + +static void __exit ip_vs_cleanup(void) +{ + nf_unregister_hooks(ip_vs_ops, ARRAY_SIZE(ip_vs_ops)); + ip_vs_conn_cleanup(); + ip_vs_app_cleanup(); + ip_vs_protocol_cleanup(); + ip_vs_control_cleanup(); + ip_vs_estimator_cleanup(); + IP_VS_INFO("ipvs unloaded.\n"); +} + +module_init(ip_vs_init); +module_exit(ip_vs_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c new file mode 100644 index 00000000000..0302cf3e503 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -0,0 +1,3443 @@ +/* + * IPVS An implementation of the IP virtual server support for the + * LINUX operating system. IPVS is now implemented as a module + * over the NetFilter framework. IPVS can be used to build a + * high-performance and highly available server based on a + * cluster of servers. + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Peter Kese <peter.kese@ijs.si> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/types.h> +#include <linux/capability.h> +#include <linux/fs.h> +#include <linux/sysctl.h> +#include <linux/proc_fs.h> +#include <linux/workqueue.h> +#include <linux/swap.h> +#include <linux/seq_file.h> + +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> +#include <linux/mutex.h> + +#include <net/net_namespace.h> +#include <net/ip.h> +#ifdef CONFIG_IP_VS_IPV6 +#include <net/ipv6.h> +#include <net/ip6_route.h> +#endif +#include <net/route.h> +#include <net/sock.h> +#include <net/genetlink.h> + +#include <asm/uaccess.h> + +#include <net/ip_vs.h> + +/* semaphore for IPVS sockopts. And, [gs]etsockopt may sleep. */ +static DEFINE_MUTEX(__ip_vs_mutex); + +/* lock for service table */ +static DEFINE_RWLOCK(__ip_vs_svc_lock); + +/* lock for table with the real services */ +static DEFINE_RWLOCK(__ip_vs_rs_lock); + +/* lock for state and timeout tables */ +static DEFINE_RWLOCK(__ip_vs_securetcp_lock); + +/* lock for drop entry handling */ +static DEFINE_SPINLOCK(__ip_vs_dropentry_lock); + +/* lock for drop packet handling */ +static DEFINE_SPINLOCK(__ip_vs_droppacket_lock); + +/* 1/rate drop and drop-entry variables */ +int ip_vs_drop_rate = 0; +int ip_vs_drop_counter = 0; +static atomic_t ip_vs_dropentry = ATOMIC_INIT(0); + +/* number of virtual services */ +static int ip_vs_num_services = 0; + +/* sysctl variables */ +static int sysctl_ip_vs_drop_entry = 0; +static int sysctl_ip_vs_drop_packet = 0; +static int sysctl_ip_vs_secure_tcp = 0; +static int sysctl_ip_vs_amemthresh = 1024; +static int sysctl_ip_vs_am_droprate = 10; +int sysctl_ip_vs_cache_bypass = 0; +int sysctl_ip_vs_expire_nodest_conn = 0; +int sysctl_ip_vs_expire_quiescent_template = 0; +int sysctl_ip_vs_sync_threshold[2] = { 3, 50 }; +int sysctl_ip_vs_nat_icmp_send = 0; + + +#ifdef CONFIG_IP_VS_DEBUG +static int sysctl_ip_vs_debug_level = 0; + +int ip_vs_get_debug_level(void) +{ + return sysctl_ip_vs_debug_level; +} +#endif + +#ifdef CONFIG_IP_VS_IPV6 +/* Taken from rt6_fill_node() in net/ipv6/route.c, is there a better way? */ +static int __ip_vs_addr_is_local_v6(const struct in6_addr *addr) +{ + struct rt6_info *rt; + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip6_u = { + .daddr = *addr, + .saddr = { .s6_addr32 = {0, 0, 0, 0} }, } }, + }; + + rt = (struct rt6_info *)ip6_route_output(&init_net, NULL, &fl); + if (rt && rt->rt6i_dev && (rt->rt6i_dev->flags & IFF_LOOPBACK)) + return 1; + + return 0; +} +#endif +/* + * update_defense_level is called from keventd and from sysctl, + * so it needs to protect itself from softirqs + */ +static void update_defense_level(void) +{ + struct sysinfo i; + static int old_secure_tcp = 0; + int availmem; + int nomem; + int to_change = -1; + + /* we only count free and buffered memory (in pages) */ + si_meminfo(&i); + availmem = i.freeram + i.bufferram; + /* however in linux 2.5 the i.bufferram is total page cache size, + we need adjust it */ + /* si_swapinfo(&i); */ + /* availmem = availmem - (i.totalswap - i.freeswap); */ + + nomem = (availmem < sysctl_ip_vs_amemthresh); + + local_bh_disable(); + + /* drop_entry */ + spin_lock(&__ip_vs_dropentry_lock); + switch (sysctl_ip_vs_drop_entry) { + case 0: + atomic_set(&ip_vs_dropentry, 0); + break; + case 1: + if (nomem) { + atomic_set(&ip_vs_dropentry, 1); + sysctl_ip_vs_drop_entry = 2; + } else { + atomic_set(&ip_vs_dropentry, 0); + } + break; + case 2: + if (nomem) { + atomic_set(&ip_vs_dropentry, 1); + } else { + atomic_set(&ip_vs_dropentry, 0); + sysctl_ip_vs_drop_entry = 1; + }; + break; + case 3: + atomic_set(&ip_vs_dropentry, 1); + break; + } + spin_unlock(&__ip_vs_dropentry_lock); + + /* drop_packet */ + spin_lock(&__ip_vs_droppacket_lock); + switch (sysctl_ip_vs_drop_packet) { + case 0: + ip_vs_drop_rate = 0; + break; + case 1: + if (nomem) { + ip_vs_drop_rate = ip_vs_drop_counter + = sysctl_ip_vs_amemthresh / + (sysctl_ip_vs_amemthresh-availmem); + sysctl_ip_vs_drop_packet = 2; + } else { + ip_vs_drop_rate = 0; + } + break; + case 2: + if (nomem) { + ip_vs_drop_rate = ip_vs_drop_counter + = sysctl_ip_vs_amemthresh / + (sysctl_ip_vs_amemthresh-availmem); + } else { + ip_vs_drop_rate = 0; + sysctl_ip_vs_drop_packet = 1; + } + break; + case 3: + ip_vs_drop_rate = sysctl_ip_vs_am_droprate; + break; + } + spin_unlock(&__ip_vs_droppacket_lock); + + /* secure_tcp */ + write_lock(&__ip_vs_securetcp_lock); + switch (sysctl_ip_vs_secure_tcp) { + case 0: + if (old_secure_tcp >= 2) + to_change = 0; + break; + case 1: + if (nomem) { + if (old_secure_tcp < 2) + to_change = 1; + sysctl_ip_vs_secure_tcp = 2; + } else { + if (old_secure_tcp >= 2) + to_change = 0; + } + break; + case 2: + if (nomem) { + if (old_secure_tcp < 2) + to_change = 1; + } else { + if (old_secure_tcp >= 2) + to_change = 0; + sysctl_ip_vs_secure_tcp = 1; + } + break; + case 3: + if (old_secure_tcp < 2) + to_change = 1; + break; + } + old_secure_tcp = sysctl_ip_vs_secure_tcp; + if (to_change >= 0) + ip_vs_protocol_timeout_change(sysctl_ip_vs_secure_tcp>1); + write_unlock(&__ip_vs_securetcp_lock); + + local_bh_enable(); +} + + +/* + * Timer for checking the defense + */ +#define DEFENSE_TIMER_PERIOD 1*HZ +static void defense_work_handler(struct work_struct *work); +static DECLARE_DELAYED_WORK(defense_work, defense_work_handler); + +static void defense_work_handler(struct work_struct *work) +{ + update_defense_level(); + if (atomic_read(&ip_vs_dropentry)) + ip_vs_random_dropentry(); + + schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); +} + +int +ip_vs_use_count_inc(void) +{ + return try_module_get(THIS_MODULE); +} + +void +ip_vs_use_count_dec(void) +{ + module_put(THIS_MODULE); +} + + +/* + * Hash table: for virtual service lookups + */ +#define IP_VS_SVC_TAB_BITS 8 +#define IP_VS_SVC_TAB_SIZE (1 << IP_VS_SVC_TAB_BITS) +#define IP_VS_SVC_TAB_MASK (IP_VS_SVC_TAB_SIZE - 1) + +/* the service table hashed by <protocol, addr, port> */ +static struct list_head ip_vs_svc_table[IP_VS_SVC_TAB_SIZE]; +/* the service table hashed by fwmark */ +static struct list_head ip_vs_svc_fwm_table[IP_VS_SVC_TAB_SIZE]; + +/* + * Hash table: for real service lookups + */ +#define IP_VS_RTAB_BITS 4 +#define IP_VS_RTAB_SIZE (1 << IP_VS_RTAB_BITS) +#define IP_VS_RTAB_MASK (IP_VS_RTAB_SIZE - 1) + +static struct list_head ip_vs_rtable[IP_VS_RTAB_SIZE]; + +/* + * Trash for destinations + */ +static LIST_HEAD(ip_vs_dest_trash); + +/* + * FTP & NULL virtual service counters + */ +static atomic_t ip_vs_ftpsvc_counter = ATOMIC_INIT(0); +static atomic_t ip_vs_nullsvc_counter = ATOMIC_INIT(0); + + +/* + * Returns hash value for virtual service + */ +static __inline__ unsigned +ip_vs_svc_hashkey(int af, unsigned proto, const union nf_inet_addr *addr, + __be16 port) +{ + register unsigned porth = ntohs(port); + __be32 addr_fold = addr->ip; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + addr_fold = addr->ip6[0]^addr->ip6[1]^ + addr->ip6[2]^addr->ip6[3]; +#endif + + return (proto^ntohl(addr_fold)^(porth>>IP_VS_SVC_TAB_BITS)^porth) + & IP_VS_SVC_TAB_MASK; +} + +/* + * Returns hash value of fwmark for virtual service lookup + */ +static __inline__ unsigned ip_vs_svc_fwm_hashkey(__u32 fwmark) +{ + return fwmark & IP_VS_SVC_TAB_MASK; +} + +/* + * Hashes a service in the ip_vs_svc_table by <proto,addr,port> + * or in the ip_vs_svc_fwm_table by fwmark. + * Should be called with locked tables. + */ +static int ip_vs_svc_hash(struct ip_vs_service *svc) +{ + unsigned hash; + + if (svc->flags & IP_VS_SVC_F_HASHED) { + IP_VS_ERR("ip_vs_svc_hash(): request for already hashed, " + "called from %p\n", __builtin_return_address(0)); + return 0; + } + + if (svc->fwmark == 0) { + /* + * Hash it by <protocol,addr,port> in ip_vs_svc_table + */ + hash = ip_vs_svc_hashkey(svc->af, svc->protocol, &svc->addr, + svc->port); + list_add(&svc->s_list, &ip_vs_svc_table[hash]); + } else { + /* + * Hash it by fwmark in ip_vs_svc_fwm_table + */ + hash = ip_vs_svc_fwm_hashkey(svc->fwmark); + list_add(&svc->f_list, &ip_vs_svc_fwm_table[hash]); + } + + svc->flags |= IP_VS_SVC_F_HASHED; + /* increase its refcnt because it is referenced by the svc table */ + atomic_inc(&svc->refcnt); + return 1; +} + + +/* + * Unhashes a service from ip_vs_svc_table/ip_vs_svc_fwm_table. + * Should be called with locked tables. + */ +static int ip_vs_svc_unhash(struct ip_vs_service *svc) +{ + if (!(svc->flags & IP_VS_SVC_F_HASHED)) { + IP_VS_ERR("ip_vs_svc_unhash(): request for unhash flagged, " + "called from %p\n", __builtin_return_address(0)); + return 0; + } + + if (svc->fwmark == 0) { + /* Remove it from the ip_vs_svc_table table */ + list_del(&svc->s_list); + } else { + /* Remove it from the ip_vs_svc_fwm_table table */ + list_del(&svc->f_list); + } + + svc->flags &= ~IP_VS_SVC_F_HASHED; + atomic_dec(&svc->refcnt); + return 1; +} + + +/* + * Get service by {proto,addr,port} in the service table. + */ +static inline struct ip_vs_service * +__ip_vs_service_get(int af, __u16 protocol, const union nf_inet_addr *vaddr, + __be16 vport) +{ + unsigned hash; + struct ip_vs_service *svc; + + /* Check for "full" addressed entries */ + hash = ip_vs_svc_hashkey(af, protocol, vaddr, vport); + + list_for_each_entry(svc, &ip_vs_svc_table[hash], s_list){ + if ((svc->af == af) + && ip_vs_addr_equal(af, &svc->addr, vaddr) + && (svc->port == vport) + && (svc->protocol == protocol)) { + /* HIT */ + atomic_inc(&svc->usecnt); + return svc; + } + } + + return NULL; +} + + +/* + * Get service by {fwmark} in the service table. + */ +static inline struct ip_vs_service * +__ip_vs_svc_fwm_get(int af, __u32 fwmark) +{ + unsigned hash; + struct ip_vs_service *svc; + + /* Check for fwmark addressed entries */ + hash = ip_vs_svc_fwm_hashkey(fwmark); + + list_for_each_entry(svc, &ip_vs_svc_fwm_table[hash], f_list) { + if (svc->fwmark == fwmark && svc->af == af) { + /* HIT */ + atomic_inc(&svc->usecnt); + return svc; + } + } + + return NULL; +} + +struct ip_vs_service * +ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, + const union nf_inet_addr *vaddr, __be16 vport) +{ + struct ip_vs_service *svc; + + read_lock(&__ip_vs_svc_lock); + + /* + * Check the table hashed by fwmark first + */ + if (fwmark && (svc = __ip_vs_svc_fwm_get(af, fwmark))) + goto out; + + /* + * Check the table hashed by <protocol,addr,port> + * for "full" addressed entries + */ + svc = __ip_vs_service_get(af, protocol, vaddr, vport); + + if (svc == NULL + && protocol == IPPROTO_TCP + && atomic_read(&ip_vs_ftpsvc_counter) + && (vport == FTPDATA || ntohs(vport) >= PROT_SOCK)) { + /* + * Check if ftp service entry exists, the packet + * might belong to FTP data connections. + */ + svc = __ip_vs_service_get(af, protocol, vaddr, FTPPORT); + } + + if (svc == NULL + && atomic_read(&ip_vs_nullsvc_counter)) { + /* + * Check if the catch-all port (port zero) exists + */ + svc = __ip_vs_service_get(af, protocol, vaddr, 0); + } + + out: + read_unlock(&__ip_vs_svc_lock); + + IP_VS_DBG_BUF(9, "lookup service: fwm %u %s %s:%u %s\n", + fwmark, ip_vs_proto_name(protocol), + IP_VS_DBG_ADDR(af, vaddr), ntohs(vport), + svc ? "hit" : "not hit"); + + return svc; +} + + +static inline void +__ip_vs_bind_svc(struct ip_vs_dest *dest, struct ip_vs_service *svc) +{ + atomic_inc(&svc->refcnt); + dest->svc = svc; +} + +static inline void +__ip_vs_unbind_svc(struct ip_vs_dest *dest) +{ + struct ip_vs_service *svc = dest->svc; + + dest->svc = NULL; + if (atomic_dec_and_test(&svc->refcnt)) + kfree(svc); +} + + +/* + * Returns hash value for real service + */ +static inline unsigned ip_vs_rs_hashkey(int af, + const union nf_inet_addr *addr, + __be16 port) +{ + register unsigned porth = ntohs(port); + __be32 addr_fold = addr->ip; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + addr_fold = addr->ip6[0]^addr->ip6[1]^ + addr->ip6[2]^addr->ip6[3]; +#endif + + return (ntohl(addr_fold)^(porth>>IP_VS_RTAB_BITS)^porth) + & IP_VS_RTAB_MASK; +} + +/* + * Hashes ip_vs_dest in ip_vs_rtable by <proto,addr,port>. + * should be called with locked tables. + */ +static int ip_vs_rs_hash(struct ip_vs_dest *dest) +{ + unsigned hash; + + if (!list_empty(&dest->d_list)) { + return 0; + } + + /* + * Hash by proto,addr,port, + * which are the parameters of the real service. + */ + hash = ip_vs_rs_hashkey(dest->af, &dest->addr, dest->port); + + list_add(&dest->d_list, &ip_vs_rtable[hash]); + + return 1; +} + +/* + * UNhashes ip_vs_dest from ip_vs_rtable. + * should be called with locked tables. + */ +static int ip_vs_rs_unhash(struct ip_vs_dest *dest) +{ + /* + * Remove it from the ip_vs_rtable table. + */ + if (!list_empty(&dest->d_list)) { + list_del(&dest->d_list); + INIT_LIST_HEAD(&dest->d_list); + } + + return 1; +} + +/* + * Lookup real service by <proto,addr,port> in the real service table. + */ +struct ip_vs_dest * +ip_vs_lookup_real_service(int af, __u16 protocol, + const union nf_inet_addr *daddr, + __be16 dport) +{ + unsigned hash; + struct ip_vs_dest *dest; + + /* + * Check for "full" addressed entries + * Return the first found entry + */ + hash = ip_vs_rs_hashkey(af, daddr, dport); + + read_lock(&__ip_vs_rs_lock); + list_for_each_entry(dest, &ip_vs_rtable[hash], d_list) { + if ((dest->af == af) + && ip_vs_addr_equal(af, &dest->addr, daddr) + && (dest->port == dport) + && ((dest->protocol == protocol) || + dest->vfwmark)) { + /* HIT */ + read_unlock(&__ip_vs_rs_lock); + return dest; + } + } + read_unlock(&__ip_vs_rs_lock); + + return NULL; +} + +/* + * Lookup destination by {addr,port} in the given service + */ +static struct ip_vs_dest * +ip_vs_lookup_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, + __be16 dport) +{ + struct ip_vs_dest *dest; + + /* + * Find the destination for the given service + */ + list_for_each_entry(dest, &svc->destinations, n_list) { + if ((dest->af == svc->af) + && ip_vs_addr_equal(svc->af, &dest->addr, daddr) + && (dest->port == dport)) { + /* HIT */ + return dest; + } + } + + return NULL; +} + +/* + * Find destination by {daddr,dport,vaddr,protocol} + * Cretaed to be used in ip_vs_process_message() in + * the backup synchronization daemon. It finds the + * destination to be bound to the received connection + * on the backup. + * + * ip_vs_lookup_real_service() looked promissing, but + * seems not working as expected. + */ +struct ip_vs_dest *ip_vs_find_dest(int af, const union nf_inet_addr *daddr, + __be16 dport, + const union nf_inet_addr *vaddr, + __be16 vport, __u16 protocol) +{ + struct ip_vs_dest *dest; + struct ip_vs_service *svc; + + svc = ip_vs_service_get(af, 0, protocol, vaddr, vport); + if (!svc) + return NULL; + dest = ip_vs_lookup_dest(svc, daddr, dport); + if (dest) + atomic_inc(&dest->refcnt); + ip_vs_service_put(svc); + return dest; +} + +/* + * Lookup dest by {svc,addr,port} in the destination trash. + * The destination trash is used to hold the destinations that are removed + * from the service table but are still referenced by some conn entries. + * The reason to add the destination trash is when the dest is temporary + * down (either by administrator or by monitor program), the dest can be + * picked back from the trash, the remaining connections to the dest can + * continue, and the counting information of the dest is also useful for + * scheduling. + */ +static struct ip_vs_dest * +ip_vs_trash_get_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, + __be16 dport) +{ + struct ip_vs_dest *dest, *nxt; + + /* + * Find the destination in trash + */ + list_for_each_entry_safe(dest, nxt, &ip_vs_dest_trash, n_list) { + IP_VS_DBG_BUF(3, "Destination %u/%s:%u still in trash, " + "dest->refcnt=%d\n", + dest->vfwmark, + IP_VS_DBG_ADDR(svc->af, &dest->addr), + ntohs(dest->port), + atomic_read(&dest->refcnt)); + if (dest->af == svc->af && + ip_vs_addr_equal(svc->af, &dest->addr, daddr) && + dest->port == dport && + dest->vfwmark == svc->fwmark && + dest->protocol == svc->protocol && + (svc->fwmark || + (ip_vs_addr_equal(svc->af, &dest->vaddr, &svc->addr) && + dest->vport == svc->port))) { + /* HIT */ + return dest; + } + + /* + * Try to purge the destination from trash if not referenced + */ + if (atomic_read(&dest->refcnt) == 1) { + IP_VS_DBG_BUF(3, "Removing destination %u/%s:%u " + "from trash\n", + dest->vfwmark, + IP_VS_DBG_ADDR(svc->af, &dest->addr), + ntohs(dest->port)); + list_del(&dest->n_list); + ip_vs_dst_reset(dest); + __ip_vs_unbind_svc(dest); + kfree(dest); + } + } + + return NULL; +} + + +/* + * Clean up all the destinations in the trash + * Called by the ip_vs_control_cleanup() + * + * When the ip_vs_control_clearup is activated by ipvs module exit, + * the service tables must have been flushed and all the connections + * are expired, and the refcnt of each destination in the trash must + * be 1, so we simply release them here. + */ +static void ip_vs_trash_cleanup(void) +{ + struct ip_vs_dest *dest, *nxt; + + list_for_each_entry_safe(dest, nxt, &ip_vs_dest_trash, n_list) { + list_del(&dest->n_list); + ip_vs_dst_reset(dest); + __ip_vs_unbind_svc(dest); + kfree(dest); + } +} + + +static void +ip_vs_zero_stats(struct ip_vs_stats *stats) +{ + spin_lock_bh(&stats->lock); + + memset(&stats->ustats, 0, sizeof(stats->ustats)); + ip_vs_zero_estimator(stats); + + spin_unlock_bh(&stats->lock); +} + +/* + * Update a destination in the given service + */ +static void +__ip_vs_update_dest(struct ip_vs_service *svc, + struct ip_vs_dest *dest, struct ip_vs_dest_user_kern *udest) +{ + int conn_flags; + + /* set the weight and the flags */ + atomic_set(&dest->weight, udest->weight); + conn_flags = udest->conn_flags | IP_VS_CONN_F_INACTIVE; + + /* check if local node and update the flags */ +#ifdef CONFIG_IP_VS_IPV6 + if (svc->af == AF_INET6) { + if (__ip_vs_addr_is_local_v6(&udest->addr.in6)) { + conn_flags = (conn_flags & ~IP_VS_CONN_F_FWD_MASK) + | IP_VS_CONN_F_LOCALNODE; + } + } else +#endif + if (inet_addr_type(&init_net, udest->addr.ip) == RTN_LOCAL) { + conn_flags = (conn_flags & ~IP_VS_CONN_F_FWD_MASK) + | IP_VS_CONN_F_LOCALNODE; + } + + /* set the IP_VS_CONN_F_NOOUTPUT flag if not masquerading/NAT */ + if ((conn_flags & IP_VS_CONN_F_FWD_MASK) != 0) { + conn_flags |= IP_VS_CONN_F_NOOUTPUT; + } else { + /* + * Put the real service in ip_vs_rtable if not present. + * For now only for NAT! + */ + write_lock_bh(&__ip_vs_rs_lock); + ip_vs_rs_hash(dest); + write_unlock_bh(&__ip_vs_rs_lock); + } + atomic_set(&dest->conn_flags, conn_flags); + + /* bind the service */ + if (!dest->svc) { + __ip_vs_bind_svc(dest, svc); + } else { + if (dest->svc != svc) { + __ip_vs_unbind_svc(dest); + ip_vs_zero_stats(&dest->stats); + __ip_vs_bind_svc(dest, svc); + } + } + + /* set the dest status flags */ + dest->flags |= IP_VS_DEST_F_AVAILABLE; + + if (udest->u_threshold == 0 || udest->u_threshold > dest->u_threshold) + dest->flags &= ~IP_VS_DEST_F_OVERLOAD; + dest->u_threshold = udest->u_threshold; + dest->l_threshold = udest->l_threshold; +} + + +/* + * Create a destination for the given service + */ +static int +ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest, + struct ip_vs_dest **dest_p) +{ + struct ip_vs_dest *dest; + unsigned atype; + + EnterFunction(2); + +#ifdef CONFIG_IP_VS_IPV6 + if (svc->af == AF_INET6) { + atype = ipv6_addr_type(&udest->addr.in6); + if ((!(atype & IPV6_ADDR_UNICAST) || + atype & IPV6_ADDR_LINKLOCAL) && + !__ip_vs_addr_is_local_v6(&udest->addr.in6)) + return -EINVAL; + } else +#endif + { + atype = inet_addr_type(&init_net, udest->addr.ip); + if (atype != RTN_LOCAL && atype != RTN_UNICAST) + return -EINVAL; + } + + dest = kzalloc(sizeof(struct ip_vs_dest), GFP_ATOMIC); + if (dest == NULL) { + IP_VS_ERR("ip_vs_new_dest: kmalloc failed.\n"); + return -ENOMEM; + } + + dest->af = svc->af; + dest->protocol = svc->protocol; + dest->vaddr = svc->addr; + dest->vport = svc->port; + dest->vfwmark = svc->fwmark; + ip_vs_addr_copy(svc->af, &dest->addr, &udest->addr); + dest->port = udest->port; + + atomic_set(&dest->activeconns, 0); + atomic_set(&dest->inactconns, 0); + atomic_set(&dest->persistconns, 0); + atomic_set(&dest->refcnt, 0); + + INIT_LIST_HEAD(&dest->d_list); + spin_lock_init(&dest->dst_lock); + spin_lock_init(&dest->stats.lock); + __ip_vs_update_dest(svc, dest, udest); + ip_vs_new_estimator(&dest->stats); + + *dest_p = dest; + + LeaveFunction(2); + return 0; +} + + +/* + * Add a destination into an existing service + */ +static int +ip_vs_add_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) +{ + struct ip_vs_dest *dest; + union nf_inet_addr daddr; + __be16 dport = udest->port; + int ret; + + EnterFunction(2); + + if (udest->weight < 0) { + IP_VS_ERR("ip_vs_add_dest(): server weight less than zero\n"); + return -ERANGE; + } + + if (udest->l_threshold > udest->u_threshold) { + IP_VS_ERR("ip_vs_add_dest(): lower threshold is higher than " + "upper threshold\n"); + return -ERANGE; + } + + ip_vs_addr_copy(svc->af, &daddr, &udest->addr); + + /* + * Check if the dest already exists in the list + */ + dest = ip_vs_lookup_dest(svc, &daddr, dport); + + if (dest != NULL) { + IP_VS_DBG(1, "ip_vs_add_dest(): dest already exists\n"); + return -EEXIST; + } + + /* + * Check if the dest already exists in the trash and + * is from the same service + */ + dest = ip_vs_trash_get_dest(svc, &daddr, dport); + + if (dest != NULL) { + IP_VS_DBG_BUF(3, "Get destination %s:%u from trash, " + "dest->refcnt=%d, service %u/%s:%u\n", + IP_VS_DBG_ADDR(svc->af, &daddr), ntohs(dport), + atomic_read(&dest->refcnt), + dest->vfwmark, + IP_VS_DBG_ADDR(svc->af, &dest->vaddr), + ntohs(dest->vport)); + + __ip_vs_update_dest(svc, dest, udest); + + /* + * Get the destination from the trash + */ + list_del(&dest->n_list); + + ip_vs_new_estimator(&dest->stats); + + write_lock_bh(&__ip_vs_svc_lock); + + /* + * Wait until all other svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + + list_add(&dest->n_list, &svc->destinations); + svc->num_dests++; + + /* call the update_service function of its scheduler */ + if (svc->scheduler->update_service) + svc->scheduler->update_service(svc); + + write_unlock_bh(&__ip_vs_svc_lock); + return 0; + } + + /* + * Allocate and initialize the dest structure + */ + ret = ip_vs_new_dest(svc, udest, &dest); + if (ret) { + return ret; + } + + /* + * Add the dest entry into the list + */ + atomic_inc(&dest->refcnt); + + write_lock_bh(&__ip_vs_svc_lock); + + /* + * Wait until all other svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + + list_add(&dest->n_list, &svc->destinations); + svc->num_dests++; + + /* call the update_service function of its scheduler */ + if (svc->scheduler->update_service) + svc->scheduler->update_service(svc); + + write_unlock_bh(&__ip_vs_svc_lock); + + LeaveFunction(2); + + return 0; +} + + +/* + * Edit a destination in the given service + */ +static int +ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) +{ + struct ip_vs_dest *dest; + union nf_inet_addr daddr; + __be16 dport = udest->port; + + EnterFunction(2); + + if (udest->weight < 0) { + IP_VS_ERR("ip_vs_edit_dest(): server weight less than zero\n"); + return -ERANGE; + } + + if (udest->l_threshold > udest->u_threshold) { + IP_VS_ERR("ip_vs_edit_dest(): lower threshold is higher than " + "upper threshold\n"); + return -ERANGE; + } + + ip_vs_addr_copy(svc->af, &daddr, &udest->addr); + + /* + * Lookup the destination list + */ + dest = ip_vs_lookup_dest(svc, &daddr, dport); + + if (dest == NULL) { + IP_VS_DBG(1, "ip_vs_edit_dest(): dest doesn't exist\n"); + return -ENOENT; + } + + __ip_vs_update_dest(svc, dest, udest); + + write_lock_bh(&__ip_vs_svc_lock); + + /* Wait until all other svc users go away */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + + /* call the update_service, because server weight may be changed */ + if (svc->scheduler->update_service) + svc->scheduler->update_service(svc); + + write_unlock_bh(&__ip_vs_svc_lock); + + LeaveFunction(2); + + return 0; +} + + +/* + * Delete a destination (must be already unlinked from the service) + */ +static void __ip_vs_del_dest(struct ip_vs_dest *dest) +{ + ip_vs_kill_estimator(&dest->stats); + + /* + * Remove it from the d-linked list with the real services. + */ + write_lock_bh(&__ip_vs_rs_lock); + ip_vs_rs_unhash(dest); + write_unlock_bh(&__ip_vs_rs_lock); + + /* + * Decrease the refcnt of the dest, and free the dest + * if nobody refers to it (refcnt=0). Otherwise, throw + * the destination into the trash. + */ + if (atomic_dec_and_test(&dest->refcnt)) { + ip_vs_dst_reset(dest); + /* simply decrease svc->refcnt here, let the caller check + and release the service if nobody refers to it. + Only user context can release destination and service, + and only one user context can update virtual service at a + time, so the operation here is OK */ + atomic_dec(&dest->svc->refcnt); + kfree(dest); + } else { + IP_VS_DBG_BUF(3, "Moving dest %s:%u into trash, " + "dest->refcnt=%d\n", + IP_VS_DBG_ADDR(dest->af, &dest->addr), + ntohs(dest->port), + atomic_read(&dest->refcnt)); + list_add(&dest->n_list, &ip_vs_dest_trash); + atomic_inc(&dest->refcnt); + } +} + + +/* + * Unlink a destination from the given service + */ +static void __ip_vs_unlink_dest(struct ip_vs_service *svc, + struct ip_vs_dest *dest, + int svcupd) +{ + dest->flags &= ~IP_VS_DEST_F_AVAILABLE; + + /* + * Remove it from the d-linked destination list. + */ + list_del(&dest->n_list); + svc->num_dests--; + + /* + * Call the update_service function of its scheduler + */ + if (svcupd && svc->scheduler->update_service) + svc->scheduler->update_service(svc); +} + + +/* + * Delete a destination server in the given service + */ +static int +ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) +{ + struct ip_vs_dest *dest; + __be16 dport = udest->port; + + EnterFunction(2); + + dest = ip_vs_lookup_dest(svc, &udest->addr, dport); + + if (dest == NULL) { + IP_VS_DBG(1, "ip_vs_del_dest(): destination not found!\n"); + return -ENOENT; + } + + write_lock_bh(&__ip_vs_svc_lock); + + /* + * Wait until all other svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + + /* + * Unlink dest from the service + */ + __ip_vs_unlink_dest(svc, dest, 1); + + write_unlock_bh(&__ip_vs_svc_lock); + + /* + * Delete the destination + */ + __ip_vs_del_dest(dest); + + LeaveFunction(2); + + return 0; +} + + +/* + * Add a service into the service hash table + */ +static int +ip_vs_add_service(struct ip_vs_service_user_kern *u, + struct ip_vs_service **svc_p) +{ + int ret = 0; + struct ip_vs_scheduler *sched = NULL; + struct ip_vs_service *svc = NULL; + + /* increase the module use count */ + ip_vs_use_count_inc(); + + /* Lookup the scheduler by 'u->sched_name' */ + sched = ip_vs_scheduler_get(u->sched_name); + if (sched == NULL) { + IP_VS_INFO("Scheduler module ip_vs_%s not found\n", + u->sched_name); + ret = -ENOENT; + goto out_mod_dec; + } + +#ifdef CONFIG_IP_VS_IPV6 + if (u->af == AF_INET6) { + if (!sched->supports_ipv6) { + ret = -EAFNOSUPPORT; + goto out_err; + } + if ((u->netmask < 1) || (u->netmask > 128)) { + ret = -EINVAL; + goto out_err; + } + } +#endif + + svc = kzalloc(sizeof(struct ip_vs_service), GFP_ATOMIC); + if (svc == NULL) { + IP_VS_DBG(1, "ip_vs_add_service: kmalloc failed.\n"); + ret = -ENOMEM; + goto out_err; + } + + /* I'm the first user of the service */ + atomic_set(&svc->usecnt, 1); + atomic_set(&svc->refcnt, 0); + + svc->af = u->af; + svc->protocol = u->protocol; + ip_vs_addr_copy(svc->af, &svc->addr, &u->addr); + svc->port = u->port; + svc->fwmark = u->fwmark; + svc->flags = u->flags; + svc->timeout = u->timeout * HZ; + svc->netmask = u->netmask; + + INIT_LIST_HEAD(&svc->destinations); + rwlock_init(&svc->sched_lock); + spin_lock_init(&svc->stats.lock); + + /* Bind the scheduler */ + ret = ip_vs_bind_scheduler(svc, sched); + if (ret) + goto out_err; + sched = NULL; + + /* Update the virtual service counters */ + if (svc->port == FTPPORT) + atomic_inc(&ip_vs_ftpsvc_counter); + else if (svc->port == 0) + atomic_inc(&ip_vs_nullsvc_counter); + + ip_vs_new_estimator(&svc->stats); + + /* Count only IPv4 services for old get/setsockopt interface */ + if (svc->af == AF_INET) + ip_vs_num_services++; + + /* Hash the service into the service table */ + write_lock_bh(&__ip_vs_svc_lock); + ip_vs_svc_hash(svc); + write_unlock_bh(&__ip_vs_svc_lock); + + *svc_p = svc; + return 0; + + out_err: + if (svc != NULL) { + if (svc->scheduler) + ip_vs_unbind_scheduler(svc); + if (svc->inc) { + local_bh_disable(); + ip_vs_app_inc_put(svc->inc); + local_bh_enable(); + } + kfree(svc); + } + ip_vs_scheduler_put(sched); + + out_mod_dec: + /* decrease the module use count */ + ip_vs_use_count_dec(); + + return ret; +} + + +/* + * Edit a service and bind it with a new scheduler + */ +static int +ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) +{ + struct ip_vs_scheduler *sched, *old_sched; + int ret = 0; + + /* + * Lookup the scheduler, by 'u->sched_name' + */ + sched = ip_vs_scheduler_get(u->sched_name); + if (sched == NULL) { + IP_VS_INFO("Scheduler module ip_vs_%s not found\n", + u->sched_name); + return -ENOENT; + } + old_sched = sched; + +#ifdef CONFIG_IP_VS_IPV6 + if (u->af == AF_INET6) { + if (!sched->supports_ipv6) { + ret = -EAFNOSUPPORT; + goto out; + } + if ((u->netmask < 1) || (u->netmask > 128)) { + ret = -EINVAL; + goto out; + } + } +#endif + + write_lock_bh(&__ip_vs_svc_lock); + + /* + * Wait until all other svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + + /* + * Set the flags and timeout value + */ + svc->flags = u->flags | IP_VS_SVC_F_HASHED; + svc->timeout = u->timeout * HZ; + svc->netmask = u->netmask; + + old_sched = svc->scheduler; + if (sched != old_sched) { + /* + * Unbind the old scheduler + */ + if ((ret = ip_vs_unbind_scheduler(svc))) { + old_sched = sched; + goto out_unlock; + } + + /* + * Bind the new scheduler + */ + if ((ret = ip_vs_bind_scheduler(svc, sched))) { + /* + * If ip_vs_bind_scheduler fails, restore the old + * scheduler. + * The main reason of failure is out of memory. + * + * The question is if the old scheduler can be + * restored all the time. TODO: if it cannot be + * restored some time, we must delete the service, + * otherwise the system may crash. + */ + ip_vs_bind_scheduler(svc, old_sched); + old_sched = sched; + goto out_unlock; + } + } + + out_unlock: + write_unlock_bh(&__ip_vs_svc_lock); +#ifdef CONFIG_IP_VS_IPV6 + out: +#endif + + if (old_sched) + ip_vs_scheduler_put(old_sched); + + return ret; +} + + +/* + * Delete a service from the service list + * - The service must be unlinked, unlocked and not referenced! + * - We are called under _bh lock + */ +static void __ip_vs_del_service(struct ip_vs_service *svc) +{ + struct ip_vs_dest *dest, *nxt; + struct ip_vs_scheduler *old_sched; + + /* Count only IPv4 services for old get/setsockopt interface */ + if (svc->af == AF_INET) + ip_vs_num_services--; + + ip_vs_kill_estimator(&svc->stats); + + /* Unbind scheduler */ + old_sched = svc->scheduler; + ip_vs_unbind_scheduler(svc); + if (old_sched) + ip_vs_scheduler_put(old_sched); + + /* Unbind app inc */ + if (svc->inc) { + ip_vs_app_inc_put(svc->inc); + svc->inc = NULL; + } + + /* + * Unlink the whole destination list + */ + list_for_each_entry_safe(dest, nxt, &svc->destinations, n_list) { + __ip_vs_unlink_dest(svc, dest, 0); + __ip_vs_del_dest(dest); + } + + /* + * Update the virtual service counters + */ + if (svc->port == FTPPORT) + atomic_dec(&ip_vs_ftpsvc_counter); + else if (svc->port == 0) + atomic_dec(&ip_vs_nullsvc_counter); + + /* + * Free the service if nobody refers to it + */ + if (atomic_read(&svc->refcnt) == 0) + kfree(svc); + + /* decrease the module use count */ + ip_vs_use_count_dec(); +} + +/* + * Delete a service from the service list + */ +static int ip_vs_del_service(struct ip_vs_service *svc) +{ + if (svc == NULL) + return -EEXIST; + + /* + * Unhash it from the service table + */ + write_lock_bh(&__ip_vs_svc_lock); + + ip_vs_svc_unhash(svc); + + /* + * Wait until all the svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 1); + + __ip_vs_del_service(svc); + + write_unlock_bh(&__ip_vs_svc_lock); + + return 0; +} + + +/* + * Flush all the virtual services + */ +static int ip_vs_flush(void) +{ + int idx; + struct ip_vs_service *svc, *nxt; + + /* + * Flush the service table hashed by <protocol,addr,port> + */ + for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry_safe(svc, nxt, &ip_vs_svc_table[idx], s_list) { + write_lock_bh(&__ip_vs_svc_lock); + ip_vs_svc_unhash(svc); + /* + * Wait until all the svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); + __ip_vs_del_service(svc); + write_unlock_bh(&__ip_vs_svc_lock); + } + } + + /* + * Flush the service table hashed by fwmark + */ + for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry_safe(svc, nxt, + &ip_vs_svc_fwm_table[idx], f_list) { + write_lock_bh(&__ip_vs_svc_lock); + ip_vs_svc_unhash(svc); + /* + * Wait until all the svc users go away. + */ + IP_VS_WAIT_WHILE(atomic_read(&svc->usecnt) > 0); + __ip_vs_del_service(svc); + write_unlock_bh(&__ip_vs_svc_lock); + } + } + + return 0; +} + + +/* + * Zero counters in a service or all services + */ +static int ip_vs_zero_service(struct ip_vs_service *svc) +{ + struct ip_vs_dest *dest; + + write_lock_bh(&__ip_vs_svc_lock); + list_for_each_entry(dest, &svc->destinations, n_list) { + ip_vs_zero_stats(&dest->stats); + } + ip_vs_zero_stats(&svc->stats); + write_unlock_bh(&__ip_vs_svc_lock); + return 0; +} + +static int ip_vs_zero_all(void) +{ + int idx; + struct ip_vs_service *svc; + + for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { + ip_vs_zero_service(svc); + } + } + + for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry(svc, &ip_vs_svc_fwm_table[idx], f_list) { + ip_vs_zero_service(svc); + } + } + + ip_vs_zero_stats(&ip_vs_stats); + return 0; +} + + +static int +proc_do_defense_mode(ctl_table *table, int write, struct file * filp, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int *valp = table->data; + int val = *valp; + int rc; + + rc = proc_dointvec(table, write, filp, buffer, lenp, ppos); + if (write && (*valp != val)) { + if ((*valp < 0) || (*valp > 3)) { + /* Restore the correct value */ + *valp = val; + } else { + update_defense_level(); + } + } + return rc; +} + + +static int +proc_do_sync_threshold(ctl_table *table, int write, struct file *filp, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int *valp = table->data; + int val[2]; + int rc; + + /* backup the value first */ + memcpy(val, valp, sizeof(val)); + + rc = proc_dointvec(table, write, filp, buffer, lenp, ppos); + if (write && (valp[0] < 0 || valp[1] < 0 || valp[0] >= valp[1])) { + /* Restore the correct value */ + memcpy(valp, val, sizeof(val)); + } + return rc; +} + + +/* + * IPVS sysctl table (under the /proc/sys/net/ipv4/vs/) + */ + +static struct ctl_table vs_vars[] = { + { + .procname = "amemthresh", + .data = &sysctl_ip_vs_amemthresh, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#ifdef CONFIG_IP_VS_DEBUG + { + .procname = "debug_level", + .data = &sysctl_ip_vs_debug_level, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif + { + .procname = "am_droprate", + .data = &sysctl_ip_vs_am_droprate, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .procname = "drop_entry", + .data = &sysctl_ip_vs_drop_entry, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_do_defense_mode, + }, + { + .procname = "drop_packet", + .data = &sysctl_ip_vs_drop_packet, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_do_defense_mode, + }, + { + .procname = "secure_tcp", + .data = &sysctl_ip_vs_secure_tcp, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_do_defense_mode, + }, +#if 0 + { + .procname = "timeout_established", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_ESTABLISHED], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_synsent", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_SYN_SENT], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_synrecv", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_SYN_RECV], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_finwait", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_FIN_WAIT], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_timewait", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_TIME_WAIT], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_close", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_CLOSE], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_closewait", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_CLOSE_WAIT], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_lastack", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_LAST_ACK], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_listen", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_LISTEN], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_synack", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_SYNACK], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_udp", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_UDP], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .procname = "timeout_icmp", + .data = &vs_timeout_table_dos.timeout[IP_VS_S_ICMP], + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, +#endif + { + .procname = "cache_bypass", + .data = &sysctl_ip_vs_cache_bypass, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .procname = "expire_nodest_conn", + .data = &sysctl_ip_vs_expire_nodest_conn, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .procname = "expire_quiescent_template", + .data = &sysctl_ip_vs_expire_quiescent_template, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .procname = "sync_threshold", + .data = &sysctl_ip_vs_sync_threshold, + .maxlen = sizeof(sysctl_ip_vs_sync_threshold), + .mode = 0644, + .proc_handler = &proc_do_sync_threshold, + }, + { + .procname = "nat_icmp_send", + .data = &sysctl_ip_vs_nat_icmp_send, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { .ctl_name = 0 } +}; + +const struct ctl_path net_vs_ctl_path[] = { + { .procname = "net", .ctl_name = CTL_NET, }, + { .procname = "ipv4", .ctl_name = NET_IPV4, }, + { .procname = "vs", }, + { } +}; +EXPORT_SYMBOL_GPL(net_vs_ctl_path); + +static struct ctl_table_header * sysctl_header; + +#ifdef CONFIG_PROC_FS + +struct ip_vs_iter { + struct list_head *table; + int bucket; +}; + +/* + * Write the contents of the VS rule table to a PROCfs file. + * (It is kept just for backward compatibility) + */ +static inline const char *ip_vs_fwd_name(unsigned flags) +{ + switch (flags & IP_VS_CONN_F_FWD_MASK) { + case IP_VS_CONN_F_LOCALNODE: + return "Local"; + case IP_VS_CONN_F_TUNNEL: + return "Tunnel"; + case IP_VS_CONN_F_DROUTE: + return "Route"; + default: + return "Masq"; + } +} + + +/* Get the Nth entry in the two lists */ +static struct ip_vs_service *ip_vs_info_array(struct seq_file *seq, loff_t pos) +{ + struct ip_vs_iter *iter = seq->private; + int idx; + struct ip_vs_service *svc; + + /* look in hash by protocol */ + for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { + if (pos-- == 0){ + iter->table = ip_vs_svc_table; + iter->bucket = idx; + return svc; + } + } + } + + /* keep looking in fwmark */ + for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry(svc, &ip_vs_svc_fwm_table[idx], f_list) { + if (pos-- == 0) { + iter->table = ip_vs_svc_fwm_table; + iter->bucket = idx; + return svc; + } + } + } + + return NULL; +} + +static void *ip_vs_info_seq_start(struct seq_file *seq, loff_t *pos) +__acquires(__ip_vs_svc_lock) +{ + + read_lock_bh(&__ip_vs_svc_lock); + return *pos ? ip_vs_info_array(seq, *pos - 1) : SEQ_START_TOKEN; +} + + +static void *ip_vs_info_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct list_head *e; + struct ip_vs_iter *iter; + struct ip_vs_service *svc; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ip_vs_info_array(seq,0); + + svc = v; + iter = seq->private; + + if (iter->table == ip_vs_svc_table) { + /* next service in table hashed by protocol */ + if ((e = svc->s_list.next) != &ip_vs_svc_table[iter->bucket]) + return list_entry(e, struct ip_vs_service, s_list); + + + while (++iter->bucket < IP_VS_SVC_TAB_SIZE) { + list_for_each_entry(svc,&ip_vs_svc_table[iter->bucket], + s_list) { + return svc; + } + } + + iter->table = ip_vs_svc_fwm_table; + iter->bucket = -1; + goto scan_fwmark; + } + + /* next service in hashed by fwmark */ + if ((e = svc->f_list.next) != &ip_vs_svc_fwm_table[iter->bucket]) + return list_entry(e, struct ip_vs_service, f_list); + + scan_fwmark: + while (++iter->bucket < IP_VS_SVC_TAB_SIZE) { + list_for_each_entry(svc, &ip_vs_svc_fwm_table[iter->bucket], + f_list) + return svc; + } + + return NULL; +} + +static void ip_vs_info_seq_stop(struct seq_file *seq, void *v) +__releases(__ip_vs_svc_lock) +{ + read_unlock_bh(&__ip_vs_svc_lock); +} + + +static int ip_vs_info_seq_show(struct seq_file *seq, void *v) +{ + if (v == SEQ_START_TOKEN) { + seq_printf(seq, + "IP Virtual Server version %d.%d.%d (size=%d)\n", + NVERSION(IP_VS_VERSION_CODE), IP_VS_CONN_TAB_SIZE); + seq_puts(seq, + "Prot LocalAddress:Port Scheduler Flags\n"); + seq_puts(seq, + " -> RemoteAddress:Port Forward Weight ActiveConn InActConn\n"); + } else { + const struct ip_vs_service *svc = v; + const struct ip_vs_iter *iter = seq->private; + const struct ip_vs_dest *dest; + + if (iter->table == ip_vs_svc_table) { +#ifdef CONFIG_IP_VS_IPV6 + if (svc->af == AF_INET6) + seq_printf(seq, "%s [" NIP6_FMT "]:%04X %s ", + ip_vs_proto_name(svc->protocol), + NIP6(svc->addr.in6), + ntohs(svc->port), + svc->scheduler->name); + else +#endif + seq_printf(seq, "%s %08X:%04X %s ", + ip_vs_proto_name(svc->protocol), + ntohl(svc->addr.ip), + ntohs(svc->port), + svc->scheduler->name); + } else { + seq_printf(seq, "FWM %08X %s ", + svc->fwmark, svc->scheduler->name); + } + + if (svc->flags & IP_VS_SVC_F_PERSISTENT) + seq_printf(seq, "persistent %d %08X\n", + svc->timeout, + ntohl(svc->netmask)); + else + seq_putc(seq, '\n'); + + list_for_each_entry(dest, &svc->destinations, n_list) { +#ifdef CONFIG_IP_VS_IPV6 + if (dest->af == AF_INET6) + seq_printf(seq, + " -> [" NIP6_FMT "]:%04X" + " %-7s %-6d %-10d %-10d\n", + NIP6(dest->addr.in6), + ntohs(dest->port), + ip_vs_fwd_name(atomic_read(&dest->conn_flags)), + atomic_read(&dest->weight), + atomic_read(&dest->activeconns), + atomic_read(&dest->inactconns)); + else +#endif + seq_printf(seq, + " -> %08X:%04X " + "%-7s %-6d %-10d %-10d\n", + ntohl(dest->addr.ip), + ntohs(dest->port), + ip_vs_fwd_name(atomic_read(&dest->conn_flags)), + atomic_read(&dest->weight), + atomic_read(&dest->activeconns), + atomic_read(&dest->inactconns)); + + } + } + return 0; +} + +static const struct seq_operations ip_vs_info_seq_ops = { + .start = ip_vs_info_seq_start, + .next = ip_vs_info_seq_next, + .stop = ip_vs_info_seq_stop, + .show = ip_vs_info_seq_show, +}; + +static int ip_vs_info_open(struct inode *inode, struct file *file) +{ + return seq_open_private(file, &ip_vs_info_seq_ops, + sizeof(struct ip_vs_iter)); +} + +static const struct file_operations ip_vs_info_fops = { + .owner = THIS_MODULE, + .open = ip_vs_info_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +#endif + +struct ip_vs_stats ip_vs_stats = { + .lock = __SPIN_LOCK_UNLOCKED(ip_vs_stats.lock), +}; + +#ifdef CONFIG_PROC_FS +static int ip_vs_stats_show(struct seq_file *seq, void *v) +{ + +/* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ + seq_puts(seq, + " Total Incoming Outgoing Incoming Outgoing\n"); + seq_printf(seq, + " Conns Packets Packets Bytes Bytes\n"); + + spin_lock_bh(&ip_vs_stats.lock); + seq_printf(seq, "%8X %8X %8X %16LX %16LX\n\n", ip_vs_stats.ustats.conns, + ip_vs_stats.ustats.inpkts, ip_vs_stats.ustats.outpkts, + (unsigned long long) ip_vs_stats.ustats.inbytes, + (unsigned long long) ip_vs_stats.ustats.outbytes); + +/* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ + seq_puts(seq, + " Conns/s Pkts/s Pkts/s Bytes/s Bytes/s\n"); + seq_printf(seq,"%8X %8X %8X %16X %16X\n", + ip_vs_stats.ustats.cps, + ip_vs_stats.ustats.inpps, + ip_vs_stats.ustats.outpps, + ip_vs_stats.ustats.inbps, + ip_vs_stats.ustats.outbps); + spin_unlock_bh(&ip_vs_stats.lock); + + return 0; +} + +static int ip_vs_stats_seq_open(struct inode *inode, struct file *file) +{ + return single_open(file, ip_vs_stats_show, NULL); +} + +static const struct file_operations ip_vs_stats_fops = { + .owner = THIS_MODULE, + .open = ip_vs_stats_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +#endif + +/* + * Set timeout values for tcp tcpfin udp in the timeout_table. + */ +static int ip_vs_set_timeout(struct ip_vs_timeout_user *u) +{ + IP_VS_DBG(2, "Setting timeout tcp:%d tcpfin:%d udp:%d\n", + u->tcp_timeout, + u->tcp_fin_timeout, + u->udp_timeout); + +#ifdef CONFIG_IP_VS_PROTO_TCP + if (u->tcp_timeout) { + ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_ESTABLISHED] + = u->tcp_timeout * HZ; + } + + if (u->tcp_fin_timeout) { + ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_FIN_WAIT] + = u->tcp_fin_timeout * HZ; + } +#endif + +#ifdef CONFIG_IP_VS_PROTO_UDP + if (u->udp_timeout) { + ip_vs_protocol_udp.timeout_table[IP_VS_UDP_S_NORMAL] + = u->udp_timeout * HZ; + } +#endif + return 0; +} + + +#define SET_CMDID(cmd) (cmd - IP_VS_BASE_CTL) +#define SERVICE_ARG_LEN (sizeof(struct ip_vs_service_user)) +#define SVCDEST_ARG_LEN (sizeof(struct ip_vs_service_user) + \ + sizeof(struct ip_vs_dest_user)) +#define TIMEOUT_ARG_LEN (sizeof(struct ip_vs_timeout_user)) +#define DAEMON_ARG_LEN (sizeof(struct ip_vs_daemon_user)) +#define MAX_ARG_LEN SVCDEST_ARG_LEN + +static const unsigned char set_arglen[SET_CMDID(IP_VS_SO_SET_MAX)+1] = { + [SET_CMDID(IP_VS_SO_SET_ADD)] = SERVICE_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_EDIT)] = SERVICE_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_DEL)] = SERVICE_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_FLUSH)] = 0, + [SET_CMDID(IP_VS_SO_SET_ADDDEST)] = SVCDEST_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_DELDEST)] = SVCDEST_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_EDITDEST)] = SVCDEST_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_TIMEOUT)] = TIMEOUT_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_STARTDAEMON)] = DAEMON_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_STOPDAEMON)] = DAEMON_ARG_LEN, + [SET_CMDID(IP_VS_SO_SET_ZERO)] = SERVICE_ARG_LEN, +}; + +static void ip_vs_copy_usvc_compat(struct ip_vs_service_user_kern *usvc, + struct ip_vs_service_user *usvc_compat) +{ + usvc->af = AF_INET; + usvc->protocol = usvc_compat->protocol; + usvc->addr.ip = usvc_compat->addr; + usvc->port = usvc_compat->port; + usvc->fwmark = usvc_compat->fwmark; + + /* Deep copy of sched_name is not needed here */ + usvc->sched_name = usvc_compat->sched_name; + + usvc->flags = usvc_compat->flags; + usvc->timeout = usvc_compat->timeout; + usvc->netmask = usvc_compat->netmask; +} + +static void ip_vs_copy_udest_compat(struct ip_vs_dest_user_kern *udest, + struct ip_vs_dest_user *udest_compat) +{ + udest->addr.ip = udest_compat->addr; + udest->port = udest_compat->port; + udest->conn_flags = udest_compat->conn_flags; + udest->weight = udest_compat->weight; + udest->u_threshold = udest_compat->u_threshold; + udest->l_threshold = udest_compat->l_threshold; +} + +static int +do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) +{ + int ret; + unsigned char arg[MAX_ARG_LEN]; + struct ip_vs_service_user *usvc_compat; + struct ip_vs_service_user_kern usvc; + struct ip_vs_service *svc; + struct ip_vs_dest_user *udest_compat; + struct ip_vs_dest_user_kern udest; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + if (len != set_arglen[SET_CMDID(cmd)]) { + IP_VS_ERR("set_ctl: len %u != %u\n", + len, set_arglen[SET_CMDID(cmd)]); + return -EINVAL; + } + + if (copy_from_user(arg, user, len) != 0) + return -EFAULT; + + /* increase the module use count */ + ip_vs_use_count_inc(); + + if (mutex_lock_interruptible(&__ip_vs_mutex)) { + ret = -ERESTARTSYS; + goto out_dec; + } + + if (cmd == IP_VS_SO_SET_FLUSH) { + /* Flush the virtual service */ + ret = ip_vs_flush(); + goto out_unlock; + } else if (cmd == IP_VS_SO_SET_TIMEOUT) { + /* Set timeout values for (tcp tcpfin udp) */ + ret = ip_vs_set_timeout((struct ip_vs_timeout_user *)arg); + goto out_unlock; + } else if (cmd == IP_VS_SO_SET_STARTDAEMON) { + struct ip_vs_daemon_user *dm = (struct ip_vs_daemon_user *)arg; + ret = start_sync_thread(dm->state, dm->mcast_ifn, dm->syncid); + goto out_unlock; + } else if (cmd == IP_VS_SO_SET_STOPDAEMON) { + struct ip_vs_daemon_user *dm = (struct ip_vs_daemon_user *)arg; + ret = stop_sync_thread(dm->state); + goto out_unlock; + } + + usvc_compat = (struct ip_vs_service_user *)arg; + udest_compat = (struct ip_vs_dest_user *)(usvc_compat + 1); + + /* We only use the new structs internally, so copy userspace compat + * structs to extended internal versions */ + ip_vs_copy_usvc_compat(&usvc, usvc_compat); + ip_vs_copy_udest_compat(&udest, udest_compat); + + if (cmd == IP_VS_SO_SET_ZERO) { + /* if no service address is set, zero counters in all */ + if (!usvc.fwmark && !usvc.addr.ip && !usvc.port) { + ret = ip_vs_zero_all(); + goto out_unlock; + } + } + + /* Check for valid protocol: TCP or UDP, even for fwmark!=0 */ + if (usvc.protocol != IPPROTO_TCP && usvc.protocol != IPPROTO_UDP) { + IP_VS_ERR("set_ctl: invalid protocol: %d %d.%d.%d.%d:%d %s\n", + usvc.protocol, NIPQUAD(usvc.addr.ip), + ntohs(usvc.port), usvc.sched_name); + ret = -EFAULT; + goto out_unlock; + } + + /* Lookup the exact service by <protocol, addr, port> or fwmark */ + if (usvc.fwmark == 0) + svc = __ip_vs_service_get(usvc.af, usvc.protocol, + &usvc.addr, usvc.port); + else + svc = __ip_vs_svc_fwm_get(usvc.af, usvc.fwmark); + + if (cmd != IP_VS_SO_SET_ADD + && (svc == NULL || svc->protocol != usvc.protocol)) { + ret = -ESRCH; + goto out_unlock; + } + + switch (cmd) { + case IP_VS_SO_SET_ADD: + if (svc != NULL) + ret = -EEXIST; + else + ret = ip_vs_add_service(&usvc, &svc); + break; + case IP_VS_SO_SET_EDIT: + ret = ip_vs_edit_service(svc, &usvc); + break; + case IP_VS_SO_SET_DEL: + ret = ip_vs_del_service(svc); + if (!ret) + goto out_unlock; + break; + case IP_VS_SO_SET_ZERO: + ret = ip_vs_zero_service(svc); + break; + case IP_VS_SO_SET_ADDDEST: + ret = ip_vs_add_dest(svc, &udest); + break; + case IP_VS_SO_SET_EDITDEST: + ret = ip_vs_edit_dest(svc, &udest); + break; + case IP_VS_SO_SET_DELDEST: + ret = ip_vs_del_dest(svc, &udest); + break; + default: + ret = -EINVAL; + } + + if (svc) + ip_vs_service_put(svc); + + out_unlock: + mutex_unlock(&__ip_vs_mutex); + out_dec: + /* decrease the module use count */ + ip_vs_use_count_dec(); + + return ret; +} + + +static void +ip_vs_copy_stats(struct ip_vs_stats_user *dst, struct ip_vs_stats *src) +{ + spin_lock_bh(&src->lock); + memcpy(dst, &src->ustats, sizeof(*dst)); + spin_unlock_bh(&src->lock); +} + +static void +ip_vs_copy_service(struct ip_vs_service_entry *dst, struct ip_vs_service *src) +{ + dst->protocol = src->protocol; + dst->addr = src->addr.ip; + dst->port = src->port; + dst->fwmark = src->fwmark; + strlcpy(dst->sched_name, src->scheduler->name, sizeof(dst->sched_name)); + dst->flags = src->flags; + dst->timeout = src->timeout / HZ; + dst->netmask = src->netmask; + dst->num_dests = src->num_dests; + ip_vs_copy_stats(&dst->stats, &src->stats); +} + +static inline int +__ip_vs_get_service_entries(const struct ip_vs_get_services *get, + struct ip_vs_get_services __user *uptr) +{ + int idx, count=0; + struct ip_vs_service *svc; + struct ip_vs_service_entry entry; + int ret = 0; + + for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { + /* Only expose IPv4 entries to old interface */ + if (svc->af != AF_INET) + continue; + + if (count >= get->num_services) + goto out; + memset(&entry, 0, sizeof(entry)); + ip_vs_copy_service(&entry, svc); + if (copy_to_user(&uptr->entrytable[count], + &entry, sizeof(entry))) { + ret = -EFAULT; + goto out; + } + count++; + } + } + + for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + list_for_each_entry(svc, &ip_vs_svc_fwm_table[idx], f_list) { + /* Only expose IPv4 entries to old interface */ + if (svc->af != AF_INET) + continue; + + if (count >= get->num_services) + goto out; + memset(&entry, 0, sizeof(entry)); + ip_vs_copy_service(&entry, svc); + if (copy_to_user(&uptr->entrytable[count], + &entry, sizeof(entry))) { + ret = -EFAULT; + goto out; + } + count++; + } + } + out: + return ret; +} + +static inline int +__ip_vs_get_dest_entries(const struct ip_vs_get_dests *get, + struct ip_vs_get_dests __user *uptr) +{ + struct ip_vs_service *svc; + union nf_inet_addr addr = { .ip = get->addr }; + int ret = 0; + + if (get->fwmark) + svc = __ip_vs_svc_fwm_get(AF_INET, get->fwmark); + else + svc = __ip_vs_service_get(AF_INET, get->protocol, &addr, + get->port); + + if (svc) { + int count = 0; + struct ip_vs_dest *dest; + struct ip_vs_dest_entry entry; + + list_for_each_entry(dest, &svc->destinations, n_list) { + if (count >= get->num_dests) + break; + + entry.addr = dest->addr.ip; + entry.port = dest->port; + entry.conn_flags = atomic_read(&dest->conn_flags); + entry.weight = atomic_read(&dest->weight); + entry.u_threshold = dest->u_threshold; + entry.l_threshold = dest->l_threshold; + entry.activeconns = atomic_read(&dest->activeconns); + entry.inactconns = atomic_read(&dest->inactconns); + entry.persistconns = atomic_read(&dest->persistconns); + ip_vs_copy_stats(&entry.stats, &dest->stats); + if (copy_to_user(&uptr->entrytable[count], + &entry, sizeof(entry))) { + ret = -EFAULT; + break; + } + count++; + } + ip_vs_service_put(svc); + } else + ret = -ESRCH; + return ret; +} + +static inline void +__ip_vs_get_timeouts(struct ip_vs_timeout_user *u) +{ +#ifdef CONFIG_IP_VS_PROTO_TCP + u->tcp_timeout = + ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_ESTABLISHED] / HZ; + u->tcp_fin_timeout = + ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_FIN_WAIT] / HZ; +#endif +#ifdef CONFIG_IP_VS_PROTO_UDP + u->udp_timeout = + ip_vs_protocol_udp.timeout_table[IP_VS_UDP_S_NORMAL] / HZ; +#endif +} + + +#define GET_CMDID(cmd) (cmd - IP_VS_BASE_CTL) +#define GET_INFO_ARG_LEN (sizeof(struct ip_vs_getinfo)) +#define GET_SERVICES_ARG_LEN (sizeof(struct ip_vs_get_services)) +#define GET_SERVICE_ARG_LEN (sizeof(struct ip_vs_service_entry)) +#define GET_DESTS_ARG_LEN (sizeof(struct ip_vs_get_dests)) +#define GET_TIMEOUT_ARG_LEN (sizeof(struct ip_vs_timeout_user)) +#define GET_DAEMON_ARG_LEN (sizeof(struct ip_vs_daemon_user) * 2) + +static const unsigned char get_arglen[GET_CMDID(IP_VS_SO_GET_MAX)+1] = { + [GET_CMDID(IP_VS_SO_GET_VERSION)] = 64, + [GET_CMDID(IP_VS_SO_GET_INFO)] = GET_INFO_ARG_LEN, + [GET_CMDID(IP_VS_SO_GET_SERVICES)] = GET_SERVICES_ARG_LEN, + [GET_CMDID(IP_VS_SO_GET_SERVICE)] = GET_SERVICE_ARG_LEN, + [GET_CMDID(IP_VS_SO_GET_DESTS)] = GET_DESTS_ARG_LEN, + [GET_CMDID(IP_VS_SO_GET_TIMEOUT)] = GET_TIMEOUT_ARG_LEN, + [GET_CMDID(IP_VS_SO_GET_DAEMON)] = GET_DAEMON_ARG_LEN, +}; + +static int +do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) +{ + unsigned char arg[128]; + int ret = 0; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + if (*len < get_arglen[GET_CMDID(cmd)]) { + IP_VS_ERR("get_ctl: len %u < %u\n", + *len, get_arglen[GET_CMDID(cmd)]); + return -EINVAL; + } + + if (copy_from_user(arg, user, get_arglen[GET_CMDID(cmd)]) != 0) + return -EFAULT; + + if (mutex_lock_interruptible(&__ip_vs_mutex)) + return -ERESTARTSYS; + + switch (cmd) { + case IP_VS_SO_GET_VERSION: + { + char buf[64]; + + sprintf(buf, "IP Virtual Server version %d.%d.%d (size=%d)", + NVERSION(IP_VS_VERSION_CODE), IP_VS_CONN_TAB_SIZE); + if (copy_to_user(user, buf, strlen(buf)+1) != 0) { + ret = -EFAULT; + goto out; + } + *len = strlen(buf)+1; + } + break; + + case IP_VS_SO_GET_INFO: + { + struct ip_vs_getinfo info; + info.version = IP_VS_VERSION_CODE; + info.size = IP_VS_CONN_TAB_SIZE; + info.num_services = ip_vs_num_services; + if (copy_to_user(user, &info, sizeof(info)) != 0) + ret = -EFAULT; + } + break; + + case IP_VS_SO_GET_SERVICES: + { + struct ip_vs_get_services *get; + int size; + + get = (struct ip_vs_get_services *)arg; + size = sizeof(*get) + + sizeof(struct ip_vs_service_entry) * get->num_services; + if (*len != size) { + IP_VS_ERR("length: %u != %u\n", *len, size); + ret = -EINVAL; + goto out; + } + ret = __ip_vs_get_service_entries(get, user); + } + break; + + case IP_VS_SO_GET_SERVICE: + { + struct ip_vs_service_entry *entry; + struct ip_vs_service *svc; + union nf_inet_addr addr; + + entry = (struct ip_vs_service_entry *)arg; + addr.ip = entry->addr; + if (entry->fwmark) + svc = __ip_vs_svc_fwm_get(AF_INET, entry->fwmark); + else + svc = __ip_vs_service_get(AF_INET, entry->protocol, + &addr, entry->port); + if (svc) { + ip_vs_copy_service(entry, svc); + if (copy_to_user(user, entry, sizeof(*entry)) != 0) + ret = -EFAULT; + ip_vs_service_put(svc); + } else + ret = -ESRCH; + } + break; + + case IP_VS_SO_GET_DESTS: + { + struct ip_vs_get_dests *get; + int size; + + get = (struct ip_vs_get_dests *)arg; + size = sizeof(*get) + + sizeof(struct ip_vs_dest_entry) * get->num_dests; + if (*len != size) { + IP_VS_ERR("length: %u != %u\n", *len, size); + ret = -EINVAL; + goto out; + } + ret = __ip_vs_get_dest_entries(get, user); + } + break; + + case IP_VS_SO_GET_TIMEOUT: + { + struct ip_vs_timeout_user t; + + __ip_vs_get_timeouts(&t); + if (copy_to_user(user, &t, sizeof(t)) != 0) + ret = -EFAULT; + } + break; + + case IP_VS_SO_GET_DAEMON: + { + struct ip_vs_daemon_user d[2]; + + memset(&d, 0, sizeof(d)); + if (ip_vs_sync_state & IP_VS_STATE_MASTER) { + d[0].state = IP_VS_STATE_MASTER; + strlcpy(d[0].mcast_ifn, ip_vs_master_mcast_ifn, sizeof(d[0].mcast_ifn)); + d[0].syncid = ip_vs_master_syncid; + } + if (ip_vs_sync_state & IP_VS_STATE_BACKUP) { + d[1].state = IP_VS_STATE_BACKUP; + strlcpy(d[1].mcast_ifn, ip_vs_backup_mcast_ifn, sizeof(d[1].mcast_ifn)); + d[1].syncid = ip_vs_backup_syncid; + } + if (copy_to_user(user, &d, sizeof(d)) != 0) + ret = -EFAULT; + } + break; + + default: + ret = -EINVAL; + } + + out: + mutex_unlock(&__ip_vs_mutex); + return ret; +} + + +static struct nf_sockopt_ops ip_vs_sockopts = { + .pf = PF_INET, + .set_optmin = IP_VS_BASE_CTL, + .set_optmax = IP_VS_SO_SET_MAX+1, + .set = do_ip_vs_set_ctl, + .get_optmin = IP_VS_BASE_CTL, + .get_optmax = IP_VS_SO_GET_MAX+1, + .get = do_ip_vs_get_ctl, + .owner = THIS_MODULE, +}; + +/* + * Generic Netlink interface + */ + +/* IPVS genetlink family */ +static struct genl_family ip_vs_genl_family = { + .id = GENL_ID_GENERATE, + .hdrsize = 0, + .name = IPVS_GENL_NAME, + .version = IPVS_GENL_VERSION, + .maxattr = IPVS_CMD_MAX, +}; + +/* Policy used for first-level command attributes */ +static const struct nla_policy ip_vs_cmd_policy[IPVS_CMD_ATTR_MAX + 1] = { + [IPVS_CMD_ATTR_SERVICE] = { .type = NLA_NESTED }, + [IPVS_CMD_ATTR_DEST] = { .type = NLA_NESTED }, + [IPVS_CMD_ATTR_DAEMON] = { .type = NLA_NESTED }, + [IPVS_CMD_ATTR_TIMEOUT_TCP] = { .type = NLA_U32 }, + [IPVS_CMD_ATTR_TIMEOUT_TCP_FIN] = { .type = NLA_U32 }, + [IPVS_CMD_ATTR_TIMEOUT_UDP] = { .type = NLA_U32 }, +}; + +/* Policy used for attributes in nested attribute IPVS_CMD_ATTR_DAEMON */ +static const struct nla_policy ip_vs_daemon_policy[IPVS_DAEMON_ATTR_MAX + 1] = { + [IPVS_DAEMON_ATTR_STATE] = { .type = NLA_U32 }, + [IPVS_DAEMON_ATTR_MCAST_IFN] = { .type = NLA_NUL_STRING, + .len = IP_VS_IFNAME_MAXLEN }, + [IPVS_DAEMON_ATTR_SYNC_ID] = { .type = NLA_U32 }, +}; + +/* Policy used for attributes in nested attribute IPVS_CMD_ATTR_SERVICE */ +static const struct nla_policy ip_vs_svc_policy[IPVS_SVC_ATTR_MAX + 1] = { + [IPVS_SVC_ATTR_AF] = { .type = NLA_U16 }, + [IPVS_SVC_ATTR_PROTOCOL] = { .type = NLA_U16 }, + [IPVS_SVC_ATTR_ADDR] = { .type = NLA_BINARY, + .len = sizeof(union nf_inet_addr) }, + [IPVS_SVC_ATTR_PORT] = { .type = NLA_U16 }, + [IPVS_SVC_ATTR_FWMARK] = { .type = NLA_U32 }, + [IPVS_SVC_ATTR_SCHED_NAME] = { .type = NLA_NUL_STRING, + .len = IP_VS_SCHEDNAME_MAXLEN }, + [IPVS_SVC_ATTR_FLAGS] = { .type = NLA_BINARY, + .len = sizeof(struct ip_vs_flags) }, + [IPVS_SVC_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPVS_SVC_ATTR_NETMASK] = { .type = NLA_U32 }, + [IPVS_SVC_ATTR_STATS] = { .type = NLA_NESTED }, +}; + +/* Policy used for attributes in nested attribute IPVS_CMD_ATTR_DEST */ +static const struct nla_policy ip_vs_dest_policy[IPVS_DEST_ATTR_MAX + 1] = { + [IPVS_DEST_ATTR_ADDR] = { .type = NLA_BINARY, + .len = sizeof(union nf_inet_addr) }, + [IPVS_DEST_ATTR_PORT] = { .type = NLA_U16 }, + [IPVS_DEST_ATTR_FWD_METHOD] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_WEIGHT] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_U_THRESH] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_L_THRESH] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_ACTIVE_CONNS] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_INACT_CONNS] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_PERSIST_CONNS] = { .type = NLA_U32 }, + [IPVS_DEST_ATTR_STATS] = { .type = NLA_NESTED }, +}; + +static int ip_vs_genl_fill_stats(struct sk_buff *skb, int container_type, + struct ip_vs_stats *stats) +{ + struct nlattr *nl_stats = nla_nest_start(skb, container_type); + if (!nl_stats) + return -EMSGSIZE; + + spin_lock_bh(&stats->lock); + + NLA_PUT_U32(skb, IPVS_STATS_ATTR_CONNS, stats->ustats.conns); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_INPKTS, stats->ustats.inpkts); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTPKTS, stats->ustats.outpkts); + NLA_PUT_U64(skb, IPVS_STATS_ATTR_INBYTES, stats->ustats.inbytes); + NLA_PUT_U64(skb, IPVS_STATS_ATTR_OUTBYTES, stats->ustats.outbytes); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_CPS, stats->ustats.cps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_INPPS, stats->ustats.inpps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTPPS, stats->ustats.outpps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_INBPS, stats->ustats.inbps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTBPS, stats->ustats.outbps); + + spin_unlock_bh(&stats->lock); + + nla_nest_end(skb, nl_stats); + + return 0; + +nla_put_failure: + spin_unlock_bh(&stats->lock); + nla_nest_cancel(skb, nl_stats); + return -EMSGSIZE; +} + +static int ip_vs_genl_fill_service(struct sk_buff *skb, + struct ip_vs_service *svc) +{ + struct nlattr *nl_service; + struct ip_vs_flags flags = { .flags = svc->flags, + .mask = ~0 }; + + nl_service = nla_nest_start(skb, IPVS_CMD_ATTR_SERVICE); + if (!nl_service) + return -EMSGSIZE; + + NLA_PUT_U16(skb, IPVS_SVC_ATTR_AF, svc->af); + + if (svc->fwmark) { + NLA_PUT_U32(skb, IPVS_SVC_ATTR_FWMARK, svc->fwmark); + } else { + NLA_PUT_U16(skb, IPVS_SVC_ATTR_PROTOCOL, svc->protocol); + NLA_PUT(skb, IPVS_SVC_ATTR_ADDR, sizeof(svc->addr), &svc->addr); + NLA_PUT_U16(skb, IPVS_SVC_ATTR_PORT, svc->port); + } + + NLA_PUT_STRING(skb, IPVS_SVC_ATTR_SCHED_NAME, svc->scheduler->name); + NLA_PUT(skb, IPVS_SVC_ATTR_FLAGS, sizeof(flags), &flags); + NLA_PUT_U32(skb, IPVS_SVC_ATTR_TIMEOUT, svc->timeout / HZ); + NLA_PUT_U32(skb, IPVS_SVC_ATTR_NETMASK, svc->netmask); + + if (ip_vs_genl_fill_stats(skb, IPVS_SVC_ATTR_STATS, &svc->stats)) + goto nla_put_failure; + + nla_nest_end(skb, nl_service); + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nl_service); + return -EMSGSIZE; +} + +static int ip_vs_genl_dump_service(struct sk_buff *skb, + struct ip_vs_service *svc, + struct netlink_callback *cb) +{ + void *hdr; + + hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, + &ip_vs_genl_family, NLM_F_MULTI, + IPVS_CMD_NEW_SERVICE); + if (!hdr) + return -EMSGSIZE; + + if (ip_vs_genl_fill_service(skb, svc) < 0) + goto nla_put_failure; + + return genlmsg_end(skb, hdr); + +nla_put_failure: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +static int ip_vs_genl_dump_services(struct sk_buff *skb, + struct netlink_callback *cb) +{ + int idx = 0, i; + int start = cb->args[0]; + struct ip_vs_service *svc; + + mutex_lock(&__ip_vs_mutex); + for (i = 0; i < IP_VS_SVC_TAB_SIZE; i++) { + list_for_each_entry(svc, &ip_vs_svc_table[i], s_list) { + if (++idx <= start) + continue; + if (ip_vs_genl_dump_service(skb, svc, cb) < 0) { + idx--; + goto nla_put_failure; + } + } + } + + for (i = 0; i < IP_VS_SVC_TAB_SIZE; i++) { + list_for_each_entry(svc, &ip_vs_svc_fwm_table[i], f_list) { + if (++idx <= start) + continue; + if (ip_vs_genl_dump_service(skb, svc, cb) < 0) { + idx--; + goto nla_put_failure; + } + } + } + +nla_put_failure: + mutex_unlock(&__ip_vs_mutex); + cb->args[0] = idx; + + return skb->len; +} + +static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, + struct nlattr *nla, int full_entry) +{ + struct nlattr *attrs[IPVS_SVC_ATTR_MAX + 1]; + struct nlattr *nla_af, *nla_port, *nla_fwmark, *nla_protocol, *nla_addr; + + /* Parse mandatory identifying service fields first */ + if (nla == NULL || + nla_parse_nested(attrs, IPVS_SVC_ATTR_MAX, nla, ip_vs_svc_policy)) + return -EINVAL; + + nla_af = attrs[IPVS_SVC_ATTR_AF]; + nla_protocol = attrs[IPVS_SVC_ATTR_PROTOCOL]; + nla_addr = attrs[IPVS_SVC_ATTR_ADDR]; + nla_port = attrs[IPVS_SVC_ATTR_PORT]; + nla_fwmark = attrs[IPVS_SVC_ATTR_FWMARK]; + + if (!(nla_af && (nla_fwmark || (nla_port && nla_protocol && nla_addr)))) + return -EINVAL; + + usvc->af = nla_get_u16(nla_af); +#ifdef CONFIG_IP_VS_IPV6 + if (usvc->af != AF_INET && usvc->af != AF_INET6) +#else + if (usvc->af != AF_INET) +#endif + return -EAFNOSUPPORT; + + if (nla_fwmark) { + usvc->protocol = IPPROTO_TCP; + usvc->fwmark = nla_get_u32(nla_fwmark); + } else { + usvc->protocol = nla_get_u16(nla_protocol); + nla_memcpy(&usvc->addr, nla_addr, sizeof(usvc->addr)); + usvc->port = nla_get_u16(nla_port); + usvc->fwmark = 0; + } + + /* If a full entry was requested, check for the additional fields */ + if (full_entry) { + struct nlattr *nla_sched, *nla_flags, *nla_timeout, + *nla_netmask; + struct ip_vs_flags flags; + struct ip_vs_service *svc; + + nla_sched = attrs[IPVS_SVC_ATTR_SCHED_NAME]; + nla_flags = attrs[IPVS_SVC_ATTR_FLAGS]; + nla_timeout = attrs[IPVS_SVC_ATTR_TIMEOUT]; + nla_netmask = attrs[IPVS_SVC_ATTR_NETMASK]; + + if (!(nla_sched && nla_flags && nla_timeout && nla_netmask)) + return -EINVAL; + + nla_memcpy(&flags, nla_flags, sizeof(flags)); + + /* prefill flags from service if it already exists */ + if (usvc->fwmark) + svc = __ip_vs_svc_fwm_get(usvc->af, usvc->fwmark); + else + svc = __ip_vs_service_get(usvc->af, usvc->protocol, + &usvc->addr, usvc->port); + if (svc) { + usvc->flags = svc->flags; + ip_vs_service_put(svc); + } else + usvc->flags = 0; + + /* set new flags from userland */ + usvc->flags = (usvc->flags & ~flags.mask) | + (flags.flags & flags.mask); + usvc->sched_name = nla_data(nla_sched); + usvc->timeout = nla_get_u32(nla_timeout); + usvc->netmask = nla_get_u32(nla_netmask); + } + + return 0; +} + +static struct ip_vs_service *ip_vs_genl_find_service(struct nlattr *nla) +{ + struct ip_vs_service_user_kern usvc; + int ret; + + ret = ip_vs_genl_parse_service(&usvc, nla, 0); + if (ret) + return ERR_PTR(ret); + + if (usvc.fwmark) + return __ip_vs_svc_fwm_get(usvc.af, usvc.fwmark); + else + return __ip_vs_service_get(usvc.af, usvc.protocol, + &usvc.addr, usvc.port); +} + +static int ip_vs_genl_fill_dest(struct sk_buff *skb, struct ip_vs_dest *dest) +{ + struct nlattr *nl_dest; + + nl_dest = nla_nest_start(skb, IPVS_CMD_ATTR_DEST); + if (!nl_dest) + return -EMSGSIZE; + + NLA_PUT(skb, IPVS_DEST_ATTR_ADDR, sizeof(dest->addr), &dest->addr); + NLA_PUT_U16(skb, IPVS_DEST_ATTR_PORT, dest->port); + + NLA_PUT_U32(skb, IPVS_DEST_ATTR_FWD_METHOD, + atomic_read(&dest->conn_flags) & IP_VS_CONN_F_FWD_MASK); + NLA_PUT_U32(skb, IPVS_DEST_ATTR_WEIGHT, atomic_read(&dest->weight)); + NLA_PUT_U32(skb, IPVS_DEST_ATTR_U_THRESH, dest->u_threshold); + NLA_PUT_U32(skb, IPVS_DEST_ATTR_L_THRESH, dest->l_threshold); + NLA_PUT_U32(skb, IPVS_DEST_ATTR_ACTIVE_CONNS, + atomic_read(&dest->activeconns)); + NLA_PUT_U32(skb, IPVS_DEST_ATTR_INACT_CONNS, + atomic_read(&dest->inactconns)); + NLA_PUT_U32(skb, IPVS_DEST_ATTR_PERSIST_CONNS, + atomic_read(&dest->persistconns)); + + if (ip_vs_genl_fill_stats(skb, IPVS_DEST_ATTR_STATS, &dest->stats)) + goto nla_put_failure; + + nla_nest_end(skb, nl_dest); + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nl_dest); + return -EMSGSIZE; +} + +static int ip_vs_genl_dump_dest(struct sk_buff *skb, struct ip_vs_dest *dest, + struct netlink_callback *cb) +{ + void *hdr; + + hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, + &ip_vs_genl_family, NLM_F_MULTI, + IPVS_CMD_NEW_DEST); + if (!hdr) + return -EMSGSIZE; + + if (ip_vs_genl_fill_dest(skb, dest) < 0) + goto nla_put_failure; + + return genlmsg_end(skb, hdr); + +nla_put_failure: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +static int ip_vs_genl_dump_dests(struct sk_buff *skb, + struct netlink_callback *cb) +{ + int idx = 0; + int start = cb->args[0]; + struct ip_vs_service *svc; + struct ip_vs_dest *dest; + struct nlattr *attrs[IPVS_CMD_ATTR_MAX + 1]; + + mutex_lock(&__ip_vs_mutex); + + /* Try to find the service for which to dump destinations */ + if (nlmsg_parse(cb->nlh, GENL_HDRLEN, attrs, + IPVS_CMD_ATTR_MAX, ip_vs_cmd_policy)) + goto out_err; + + svc = ip_vs_genl_find_service(attrs[IPVS_CMD_ATTR_SERVICE]); + if (IS_ERR(svc) || svc == NULL) + goto out_err; + + /* Dump the destinations */ + list_for_each_entry(dest, &svc->destinations, n_list) { + if (++idx <= start) + continue; + if (ip_vs_genl_dump_dest(skb, dest, cb) < 0) { + idx--; + goto nla_put_failure; + } + } + +nla_put_failure: + cb->args[0] = idx; + ip_vs_service_put(svc); + +out_err: + mutex_unlock(&__ip_vs_mutex); + + return skb->len; +} + +static int ip_vs_genl_parse_dest(struct ip_vs_dest_user_kern *udest, + struct nlattr *nla, int full_entry) +{ + struct nlattr *attrs[IPVS_DEST_ATTR_MAX + 1]; + struct nlattr *nla_addr, *nla_port; + + /* Parse mandatory identifying destination fields first */ + if (nla == NULL || + nla_parse_nested(attrs, IPVS_DEST_ATTR_MAX, nla, ip_vs_dest_policy)) + return -EINVAL; + + nla_addr = attrs[IPVS_DEST_ATTR_ADDR]; + nla_port = attrs[IPVS_DEST_ATTR_PORT]; + + if (!(nla_addr && nla_port)) + return -EINVAL; + + nla_memcpy(&udest->addr, nla_addr, sizeof(udest->addr)); + udest->port = nla_get_u16(nla_port); + + /* If a full entry was requested, check for the additional fields */ + if (full_entry) { + struct nlattr *nla_fwd, *nla_weight, *nla_u_thresh, + *nla_l_thresh; + + nla_fwd = attrs[IPVS_DEST_ATTR_FWD_METHOD]; + nla_weight = attrs[IPVS_DEST_ATTR_WEIGHT]; + nla_u_thresh = attrs[IPVS_DEST_ATTR_U_THRESH]; + nla_l_thresh = attrs[IPVS_DEST_ATTR_L_THRESH]; + + if (!(nla_fwd && nla_weight && nla_u_thresh && nla_l_thresh)) + return -EINVAL; + + udest->conn_flags = nla_get_u32(nla_fwd) + & IP_VS_CONN_F_FWD_MASK; + udest->weight = nla_get_u32(nla_weight); + udest->u_threshold = nla_get_u32(nla_u_thresh); + udest->l_threshold = nla_get_u32(nla_l_thresh); + } + + return 0; +} + +static int ip_vs_genl_fill_daemon(struct sk_buff *skb, __be32 state, + const char *mcast_ifn, __be32 syncid) +{ + struct nlattr *nl_daemon; + + nl_daemon = nla_nest_start(skb, IPVS_CMD_ATTR_DAEMON); + if (!nl_daemon) + return -EMSGSIZE; + + NLA_PUT_U32(skb, IPVS_DAEMON_ATTR_STATE, state); + NLA_PUT_STRING(skb, IPVS_DAEMON_ATTR_MCAST_IFN, mcast_ifn); + NLA_PUT_U32(skb, IPVS_DAEMON_ATTR_SYNC_ID, syncid); + + nla_nest_end(skb, nl_daemon); + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nl_daemon); + return -EMSGSIZE; +} + +static int ip_vs_genl_dump_daemon(struct sk_buff *skb, __be32 state, + const char *mcast_ifn, __be32 syncid, + struct netlink_callback *cb) +{ + void *hdr; + hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, + &ip_vs_genl_family, NLM_F_MULTI, + IPVS_CMD_NEW_DAEMON); + if (!hdr) + return -EMSGSIZE; + + if (ip_vs_genl_fill_daemon(skb, state, mcast_ifn, syncid)) + goto nla_put_failure; + + return genlmsg_end(skb, hdr); + +nla_put_failure: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +static int ip_vs_genl_dump_daemons(struct sk_buff *skb, + struct netlink_callback *cb) +{ + mutex_lock(&__ip_vs_mutex); + if ((ip_vs_sync_state & IP_VS_STATE_MASTER) && !cb->args[0]) { + if (ip_vs_genl_dump_daemon(skb, IP_VS_STATE_MASTER, + ip_vs_master_mcast_ifn, + ip_vs_master_syncid, cb) < 0) + goto nla_put_failure; + + cb->args[0] = 1; + } + + if ((ip_vs_sync_state & IP_VS_STATE_BACKUP) && !cb->args[1]) { + if (ip_vs_genl_dump_daemon(skb, IP_VS_STATE_BACKUP, + ip_vs_backup_mcast_ifn, + ip_vs_backup_syncid, cb) < 0) + goto nla_put_failure; + + cb->args[1] = 1; + } + +nla_put_failure: + mutex_unlock(&__ip_vs_mutex); + + return skb->len; +} + +static int ip_vs_genl_new_daemon(struct nlattr **attrs) +{ + if (!(attrs[IPVS_DAEMON_ATTR_STATE] && + attrs[IPVS_DAEMON_ATTR_MCAST_IFN] && + attrs[IPVS_DAEMON_ATTR_SYNC_ID])) + return -EINVAL; + + return start_sync_thread(nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE]), + nla_data(attrs[IPVS_DAEMON_ATTR_MCAST_IFN]), + nla_get_u32(attrs[IPVS_DAEMON_ATTR_SYNC_ID])); +} + +static int ip_vs_genl_del_daemon(struct nlattr **attrs) +{ + if (!attrs[IPVS_DAEMON_ATTR_STATE]) + return -EINVAL; + + return stop_sync_thread(nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE])); +} + +static int ip_vs_genl_set_config(struct nlattr **attrs) +{ + struct ip_vs_timeout_user t; + + __ip_vs_get_timeouts(&t); + + if (attrs[IPVS_CMD_ATTR_TIMEOUT_TCP]) + t.tcp_timeout = nla_get_u32(attrs[IPVS_CMD_ATTR_TIMEOUT_TCP]); + + if (attrs[IPVS_CMD_ATTR_TIMEOUT_TCP_FIN]) + t.tcp_fin_timeout = + nla_get_u32(attrs[IPVS_CMD_ATTR_TIMEOUT_TCP_FIN]); + + if (attrs[IPVS_CMD_ATTR_TIMEOUT_UDP]) + t.udp_timeout = nla_get_u32(attrs[IPVS_CMD_ATTR_TIMEOUT_UDP]); + + return ip_vs_set_timeout(&t); +} + +static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) +{ + struct ip_vs_service *svc = NULL; + struct ip_vs_service_user_kern usvc; + struct ip_vs_dest_user_kern udest; + int ret = 0, cmd; + int need_full_svc = 0, need_full_dest = 0; + + cmd = info->genlhdr->cmd; + + mutex_lock(&__ip_vs_mutex); + + if (cmd == IPVS_CMD_FLUSH) { + ret = ip_vs_flush(); + goto out; + } else if (cmd == IPVS_CMD_SET_CONFIG) { + ret = ip_vs_genl_set_config(info->attrs); + goto out; + } else if (cmd == IPVS_CMD_NEW_DAEMON || + cmd == IPVS_CMD_DEL_DAEMON) { + + struct nlattr *daemon_attrs[IPVS_DAEMON_ATTR_MAX + 1]; + + if (!info->attrs[IPVS_CMD_ATTR_DAEMON] || + nla_parse_nested(daemon_attrs, IPVS_DAEMON_ATTR_MAX, + info->attrs[IPVS_CMD_ATTR_DAEMON], + ip_vs_daemon_policy)) { + ret = -EINVAL; + goto out; + } + + if (cmd == IPVS_CMD_NEW_DAEMON) + ret = ip_vs_genl_new_daemon(daemon_attrs); + else + ret = ip_vs_genl_del_daemon(daemon_attrs); + goto out; + } else if (cmd == IPVS_CMD_ZERO && + !info->attrs[IPVS_CMD_ATTR_SERVICE]) { + ret = ip_vs_zero_all(); + goto out; + } + + /* All following commands require a service argument, so check if we + * received a valid one. We need a full service specification when + * adding / editing a service. Only identifying members otherwise. */ + if (cmd == IPVS_CMD_NEW_SERVICE || cmd == IPVS_CMD_SET_SERVICE) + need_full_svc = 1; + + ret = ip_vs_genl_parse_service(&usvc, + info->attrs[IPVS_CMD_ATTR_SERVICE], + need_full_svc); + if (ret) + goto out; + + /* Lookup the exact service by <protocol, addr, port> or fwmark */ + if (usvc.fwmark == 0) + svc = __ip_vs_service_get(usvc.af, usvc.protocol, + &usvc.addr, usvc.port); + else + svc = __ip_vs_svc_fwm_get(usvc.af, usvc.fwmark); + + /* Unless we're adding a new service, the service must already exist */ + if ((cmd != IPVS_CMD_NEW_SERVICE) && (svc == NULL)) { + ret = -ESRCH; + goto out; + } + + /* Destination commands require a valid destination argument. For + * adding / editing a destination, we need a full destination + * specification. */ + if (cmd == IPVS_CMD_NEW_DEST || cmd == IPVS_CMD_SET_DEST || + cmd == IPVS_CMD_DEL_DEST) { + if (cmd != IPVS_CMD_DEL_DEST) + need_full_dest = 1; + + ret = ip_vs_genl_parse_dest(&udest, + info->attrs[IPVS_CMD_ATTR_DEST], + need_full_dest); + if (ret) + goto out; + } + + switch (cmd) { + case IPVS_CMD_NEW_SERVICE: + if (svc == NULL) + ret = ip_vs_add_service(&usvc, &svc); + else + ret = -EEXIST; + break; + case IPVS_CMD_SET_SERVICE: + ret = ip_vs_edit_service(svc, &usvc); + break; + case IPVS_CMD_DEL_SERVICE: + ret = ip_vs_del_service(svc); + break; + case IPVS_CMD_NEW_DEST: + ret = ip_vs_add_dest(svc, &udest); + break; + case IPVS_CMD_SET_DEST: + ret = ip_vs_edit_dest(svc, &udest); + break; + case IPVS_CMD_DEL_DEST: + ret = ip_vs_del_dest(svc, &udest); + break; + case IPVS_CMD_ZERO: + ret = ip_vs_zero_service(svc); + break; + default: + ret = -EINVAL; + } + +out: + if (svc) + ip_vs_service_put(svc); + mutex_unlock(&__ip_vs_mutex); + + return ret; +} + +static int ip_vs_genl_get_cmd(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *msg; + void *reply; + int ret, cmd, reply_cmd; + + cmd = info->genlhdr->cmd; + + if (cmd == IPVS_CMD_GET_SERVICE) + reply_cmd = IPVS_CMD_NEW_SERVICE; + else if (cmd == IPVS_CMD_GET_INFO) + reply_cmd = IPVS_CMD_SET_INFO; + else if (cmd == IPVS_CMD_GET_CONFIG) + reply_cmd = IPVS_CMD_SET_CONFIG; + else { + IP_VS_ERR("unknown Generic Netlink command\n"); + return -EINVAL; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + mutex_lock(&__ip_vs_mutex); + + reply = genlmsg_put_reply(msg, info, &ip_vs_genl_family, 0, reply_cmd); + if (reply == NULL) + goto nla_put_failure; + + switch (cmd) { + case IPVS_CMD_GET_SERVICE: + { + struct ip_vs_service *svc; + + svc = ip_vs_genl_find_service(info->attrs[IPVS_CMD_ATTR_SERVICE]); + if (IS_ERR(svc)) { + ret = PTR_ERR(svc); + goto out_err; + } else if (svc) { + ret = ip_vs_genl_fill_service(msg, svc); + ip_vs_service_put(svc); + if (ret) + goto nla_put_failure; + } else { + ret = -ESRCH; + goto out_err; + } + + break; + } + + case IPVS_CMD_GET_CONFIG: + { + struct ip_vs_timeout_user t; + + __ip_vs_get_timeouts(&t); +#ifdef CONFIG_IP_VS_PROTO_TCP + NLA_PUT_U32(msg, IPVS_CMD_ATTR_TIMEOUT_TCP, t.tcp_timeout); + NLA_PUT_U32(msg, IPVS_CMD_ATTR_TIMEOUT_TCP_FIN, + t.tcp_fin_timeout); +#endif +#ifdef CONFIG_IP_VS_PROTO_UDP + NLA_PUT_U32(msg, IPVS_CMD_ATTR_TIMEOUT_UDP, t.udp_timeout); +#endif + + break; + } + + case IPVS_CMD_GET_INFO: + NLA_PUT_U32(msg, IPVS_INFO_ATTR_VERSION, IP_VS_VERSION_CODE); + NLA_PUT_U32(msg, IPVS_INFO_ATTR_CONN_TAB_SIZE, + IP_VS_CONN_TAB_SIZE); + break; + } + + genlmsg_end(msg, reply); + ret = genlmsg_unicast(msg, info->snd_pid); + goto out; + +nla_put_failure: + IP_VS_ERR("not enough space in Netlink message\n"); + ret = -EMSGSIZE; + +out_err: + nlmsg_free(msg); +out: + mutex_unlock(&__ip_vs_mutex); + + return ret; +} + + +static struct genl_ops ip_vs_genl_ops[] __read_mostly = { + { + .cmd = IPVS_CMD_NEW_SERVICE, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_SET_SERVICE, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_DEL_SERVICE, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_GET_SERVICE, + .flags = GENL_ADMIN_PERM, + .doit = ip_vs_genl_get_cmd, + .dumpit = ip_vs_genl_dump_services, + .policy = ip_vs_cmd_policy, + }, + { + .cmd = IPVS_CMD_NEW_DEST, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_SET_DEST, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_DEL_DEST, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_GET_DEST, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .dumpit = ip_vs_genl_dump_dests, + }, + { + .cmd = IPVS_CMD_NEW_DAEMON, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_DEL_DAEMON, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_GET_DAEMON, + .flags = GENL_ADMIN_PERM, + .dumpit = ip_vs_genl_dump_daemons, + }, + { + .cmd = IPVS_CMD_SET_CONFIG, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_GET_CONFIG, + .flags = GENL_ADMIN_PERM, + .doit = ip_vs_genl_get_cmd, + }, + { + .cmd = IPVS_CMD_GET_INFO, + .flags = GENL_ADMIN_PERM, + .doit = ip_vs_genl_get_cmd, + }, + { + .cmd = IPVS_CMD_ZERO, + .flags = GENL_ADMIN_PERM, + .policy = ip_vs_cmd_policy, + .doit = ip_vs_genl_set_cmd, + }, + { + .cmd = IPVS_CMD_FLUSH, + .flags = GENL_ADMIN_PERM, + .doit = ip_vs_genl_set_cmd, + }, +}; + +static int __init ip_vs_genl_register(void) +{ + int ret, i; + + ret = genl_register_family(&ip_vs_genl_family); + if (ret) + return ret; + + for (i = 0; i < ARRAY_SIZE(ip_vs_genl_ops); i++) { + ret = genl_register_ops(&ip_vs_genl_family, &ip_vs_genl_ops[i]); + if (ret) + goto err_out; + } + return 0; + +err_out: + genl_unregister_family(&ip_vs_genl_family); + return ret; +} + +static void ip_vs_genl_unregister(void) +{ + genl_unregister_family(&ip_vs_genl_family); +} + +/* End of Generic Netlink interface definitions */ + + +int __init ip_vs_control_init(void) +{ + int ret; + int idx; + + EnterFunction(2); + + ret = nf_register_sockopt(&ip_vs_sockopts); + if (ret) { + IP_VS_ERR("cannot register sockopt.\n"); + return ret; + } + + ret = ip_vs_genl_register(); + if (ret) { + IP_VS_ERR("cannot register Generic Netlink interface.\n"); + nf_unregister_sockopt(&ip_vs_sockopts); + return ret; + } + + proc_net_fops_create(&init_net, "ip_vs", 0, &ip_vs_info_fops); + proc_net_fops_create(&init_net, "ip_vs_stats",0, &ip_vs_stats_fops); + + sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars); + + /* Initialize ip_vs_svc_table, ip_vs_svc_fwm_table, ip_vs_rtable */ + for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { + INIT_LIST_HEAD(&ip_vs_svc_table[idx]); + INIT_LIST_HEAD(&ip_vs_svc_fwm_table[idx]); + } + for(idx = 0; idx < IP_VS_RTAB_SIZE; idx++) { + INIT_LIST_HEAD(&ip_vs_rtable[idx]); + } + + ip_vs_new_estimator(&ip_vs_stats); + + /* Hook the defense timer */ + schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); + + LeaveFunction(2); + return 0; +} + + +void ip_vs_control_cleanup(void) +{ + EnterFunction(2); + ip_vs_trash_cleanup(); + cancel_rearming_delayed_work(&defense_work); + cancel_work_sync(&defense_work.work); + ip_vs_kill_estimator(&ip_vs_stats); + unregister_sysctl_table(sysctl_header); + proc_net_remove(&init_net, "ip_vs_stats"); + proc_net_remove(&init_net, "ip_vs"); + ip_vs_genl_unregister(); + nf_unregister_sockopt(&ip_vs_sockopts); + LeaveFunction(2); +} diff --git a/net/netfilter/ipvs/ip_vs_dh.c b/net/netfilter/ipvs/ip_vs_dh.c new file mode 100644 index 00000000000..a16943fd72f --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_dh.c @@ -0,0 +1,261 @@ +/* + * IPVS: Destination Hashing scheduling module + * + * Authors: Wensong Zhang <wensong@gnuchina.org> + * + * Inspired by the consistent hashing scheduler patch from + * Thomas Proell <proellt@gmx.de> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +/* + * The dh algorithm is to select server by the hash key of destination IP + * address. The pseudo code is as follows: + * + * n <- servernode[dest_ip]; + * if (n is dead) OR + * (n is overloaded) OR (n.weight <= 0) then + * return NULL; + * + * return n; + * + * Notes that servernode is a 256-bucket hash table that maps the hash + * index derived from packet destination IP address to the current server + * array. If the dh scheduler is used in cache cluster, it is good to + * combine it with cache_bypass feature. When the statically assigned + * server is dead or overloaded, the load balancer can bypass the cache + * server and send requests to the original server directly. + * + */ + +#include <linux/ip.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> + +#include <net/ip_vs.h> + + +/* + * IPVS DH bucket + */ +struct ip_vs_dh_bucket { + struct ip_vs_dest *dest; /* real server (cache) */ +}; + +/* + * for IPVS DH entry hash table + */ +#ifndef CONFIG_IP_VS_DH_TAB_BITS +#define CONFIG_IP_VS_DH_TAB_BITS 8 +#endif +#define IP_VS_DH_TAB_BITS CONFIG_IP_VS_DH_TAB_BITS +#define IP_VS_DH_TAB_SIZE (1 << IP_VS_DH_TAB_BITS) +#define IP_VS_DH_TAB_MASK (IP_VS_DH_TAB_SIZE - 1) + + +/* + * Returns hash value for IPVS DH entry + */ +static inline unsigned ip_vs_dh_hashkey(__be32 addr) +{ + return (ntohl(addr)*2654435761UL) & IP_VS_DH_TAB_MASK; +} + + +/* + * Get ip_vs_dest associated with supplied parameters. + */ +static inline struct ip_vs_dest * +ip_vs_dh_get(struct ip_vs_dh_bucket *tbl, __be32 addr) +{ + return (tbl[ip_vs_dh_hashkey(addr)]).dest; +} + + +/* + * Assign all the hash buckets of the specified table with the service. + */ +static int +ip_vs_dh_assign(struct ip_vs_dh_bucket *tbl, struct ip_vs_service *svc) +{ + int i; + struct ip_vs_dh_bucket *b; + struct list_head *p; + struct ip_vs_dest *dest; + + b = tbl; + p = &svc->destinations; + for (i=0; i<IP_VS_DH_TAB_SIZE; i++) { + if (list_empty(p)) { + b->dest = NULL; + } else { + if (p == &svc->destinations) + p = p->next; + + dest = list_entry(p, struct ip_vs_dest, n_list); + atomic_inc(&dest->refcnt); + b->dest = dest; + + p = p->next; + } + b++; + } + return 0; +} + + +/* + * Flush all the hash buckets of the specified table. + */ +static void ip_vs_dh_flush(struct ip_vs_dh_bucket *tbl) +{ + int i; + struct ip_vs_dh_bucket *b; + + b = tbl; + for (i=0; i<IP_VS_DH_TAB_SIZE; i++) { + if (b->dest) { + atomic_dec(&b->dest->refcnt); + b->dest = NULL; + } + b++; + } +} + + +static int ip_vs_dh_init_svc(struct ip_vs_service *svc) +{ + struct ip_vs_dh_bucket *tbl; + + /* allocate the DH table for this service */ + tbl = kmalloc(sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE, + GFP_ATOMIC); + if (tbl == NULL) { + IP_VS_ERR("ip_vs_dh_init_svc(): no memory\n"); + return -ENOMEM; + } + svc->sched_data = tbl; + IP_VS_DBG(6, "DH hash table (memory=%Zdbytes) allocated for " + "current service\n", + sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE); + + /* assign the hash buckets with the updated service */ + ip_vs_dh_assign(tbl, svc); + + return 0; +} + + +static int ip_vs_dh_done_svc(struct ip_vs_service *svc) +{ + struct ip_vs_dh_bucket *tbl = svc->sched_data; + + /* got to clean up hash buckets here */ + ip_vs_dh_flush(tbl); + + /* release the table itself */ + kfree(svc->sched_data); + IP_VS_DBG(6, "DH hash table (memory=%Zdbytes) released\n", + sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE); + + return 0; +} + + +static int ip_vs_dh_update_svc(struct ip_vs_service *svc) +{ + struct ip_vs_dh_bucket *tbl = svc->sched_data; + + /* got to clean up hash buckets here */ + ip_vs_dh_flush(tbl); + + /* assign the hash buckets with the updated service */ + ip_vs_dh_assign(tbl, svc); + + return 0; +} + + +/* + * If the dest flags is set with IP_VS_DEST_F_OVERLOAD, + * consider that the server is overloaded here. + */ +static inline int is_overloaded(struct ip_vs_dest *dest) +{ + return dest->flags & IP_VS_DEST_F_OVERLOAD; +} + + +/* + * Destination hashing scheduling + */ +static struct ip_vs_dest * +ip_vs_dh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest; + struct ip_vs_dh_bucket *tbl; + struct iphdr *iph = ip_hdr(skb); + + IP_VS_DBG(6, "ip_vs_dh_schedule(): Scheduling...\n"); + + tbl = (struct ip_vs_dh_bucket *)svc->sched_data; + dest = ip_vs_dh_get(tbl, iph->daddr); + if (!dest + || !(dest->flags & IP_VS_DEST_F_AVAILABLE) + || atomic_read(&dest->weight) <= 0 + || is_overloaded(dest)) { + return NULL; + } + + IP_VS_DBG(6, "DH: destination IP address %u.%u.%u.%u " + "--> server %u.%u.%u.%u:%d\n", + NIPQUAD(iph->daddr), + NIPQUAD(dest->addr.ip), + ntohs(dest->port)); + + return dest; +} + + +/* + * IPVS DH Scheduler structure + */ +static struct ip_vs_scheduler ip_vs_dh_scheduler = +{ + .name = "dh", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_dh_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 0, +#endif + .init_service = ip_vs_dh_init_svc, + .done_service = ip_vs_dh_done_svc, + .update_service = ip_vs_dh_update_svc, + .schedule = ip_vs_dh_schedule, +}; + + +static int __init ip_vs_dh_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_dh_scheduler); +} + + +static void __exit ip_vs_dh_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_dh_scheduler); +} + + +module_init(ip_vs_dh_init); +module_exit(ip_vs_dh_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c new file mode 100644 index 00000000000..2eb2860dabb --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -0,0 +1,166 @@ +/* + * ip_vs_est.c: simple rate estimator for IPVS + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ +#include <linux/kernel.h> +#include <linux/jiffies.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/interrupt.h> +#include <linux/sysctl.h> +#include <linux/list.h> + +#include <net/ip_vs.h> + +/* + This code is to estimate rate in a shorter interval (such as 8 + seconds) for virtual services and real servers. For measure rate in a + long interval, it is easy to implement a user level daemon which + periodically reads those statistical counters and measure rate. + + Currently, the measurement is activated by slow timer handler. Hope + this measurement will not introduce too much load. + + We measure rate during the last 8 seconds every 2 seconds: + + avgrate = avgrate*(1-W) + rate*W + + where W = 2^(-2) + + NOTES. + + * The stored value for average bps is scaled by 2^5, so that maximal + rate is ~2.15Gbits/s, average pps and cps are scaled by 2^10. + + * A lot code is taken from net/sched/estimator.c + */ + + +static void estimation_timer(unsigned long arg); + +static LIST_HEAD(est_list); +static DEFINE_SPINLOCK(est_lock); +static DEFINE_TIMER(est_timer, estimation_timer, 0, 0); + +static void estimation_timer(unsigned long arg) +{ + struct ip_vs_estimator *e; + struct ip_vs_stats *s; + u32 n_conns; + u32 n_inpkts, n_outpkts; + u64 n_inbytes, n_outbytes; + u32 rate; + + spin_lock(&est_lock); + list_for_each_entry(e, &est_list, list) { + s = container_of(e, struct ip_vs_stats, est); + + spin_lock(&s->lock); + n_conns = s->ustats.conns; + n_inpkts = s->ustats.inpkts; + n_outpkts = s->ustats.outpkts; + n_inbytes = s->ustats.inbytes; + n_outbytes = s->ustats.outbytes; + + /* scaled by 2^10, but divided 2 seconds */ + rate = (n_conns - e->last_conns)<<9; + e->last_conns = n_conns; + e->cps += ((long)rate - (long)e->cps)>>2; + s->ustats.cps = (e->cps+0x1FF)>>10; + + rate = (n_inpkts - e->last_inpkts)<<9; + e->last_inpkts = n_inpkts; + e->inpps += ((long)rate - (long)e->inpps)>>2; + s->ustats.inpps = (e->inpps+0x1FF)>>10; + + rate = (n_outpkts - e->last_outpkts)<<9; + e->last_outpkts = n_outpkts; + e->outpps += ((long)rate - (long)e->outpps)>>2; + s->ustats.outpps = (e->outpps+0x1FF)>>10; + + rate = (n_inbytes - e->last_inbytes)<<4; + e->last_inbytes = n_inbytes; + e->inbps += ((long)rate - (long)e->inbps)>>2; + s->ustats.inbps = (e->inbps+0xF)>>5; + + rate = (n_outbytes - e->last_outbytes)<<4; + e->last_outbytes = n_outbytes; + e->outbps += ((long)rate - (long)e->outbps)>>2; + s->ustats.outbps = (e->outbps+0xF)>>5; + spin_unlock(&s->lock); + } + spin_unlock(&est_lock); + mod_timer(&est_timer, jiffies + 2*HZ); +} + +void ip_vs_new_estimator(struct ip_vs_stats *stats) +{ + struct ip_vs_estimator *est = &stats->est; + + INIT_LIST_HEAD(&est->list); + + est->last_conns = stats->ustats.conns; + est->cps = stats->ustats.cps<<10; + + est->last_inpkts = stats->ustats.inpkts; + est->inpps = stats->ustats.inpps<<10; + + est->last_outpkts = stats->ustats.outpkts; + est->outpps = stats->ustats.outpps<<10; + + est->last_inbytes = stats->ustats.inbytes; + est->inbps = stats->ustats.inbps<<5; + + est->last_outbytes = stats->ustats.outbytes; + est->outbps = stats->ustats.outbps<<5; + + spin_lock_bh(&est_lock); + list_add(&est->list, &est_list); + spin_unlock_bh(&est_lock); +} + +void ip_vs_kill_estimator(struct ip_vs_stats *stats) +{ + struct ip_vs_estimator *est = &stats->est; + + spin_lock_bh(&est_lock); + list_del(&est->list); + spin_unlock_bh(&est_lock); +} + +void ip_vs_zero_estimator(struct ip_vs_stats *stats) +{ + struct ip_vs_estimator *est = &stats->est; + + /* set counters zero, caller must hold the stats->lock lock */ + est->last_inbytes = 0; + est->last_outbytes = 0; + est->last_conns = 0; + est->last_inpkts = 0; + est->last_outpkts = 0; + est->cps = 0; + est->inpps = 0; + est->outpps = 0; + est->inbps = 0; + est->outbps = 0; +} + +int __init ip_vs_estimator_init(void) +{ + mod_timer(&est_timer, jiffies + 2 * HZ); + return 0; +} + +void ip_vs_estimator_cleanup(void) +{ + del_timer_sync(&est_timer); +} diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c new file mode 100644 index 00000000000..2e7dbd8b73a --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -0,0 +1,410 @@ +/* + * ip_vs_ftp.c: IPVS ftp application module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * Changes: + * + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Most code here is taken from ip_masq_ftp.c in kernel 2.2. The difference + * is that ip_vs_ftp module handles the reverse direction to ip_masq_ftp. + * + * IP_MASQ_FTP ftp masquerading module + * + * Version: @(#)ip_masq_ftp.c 0.04 02/05/96 + * + * Author: Wouter Gadeyne + * + */ + +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/ip.h> +#include <linux/netfilter.h> +#include <net/protocol.h> +#include <net/tcp.h> +#include <asm/unaligned.h> + +#include <net/ip_vs.h> + + +#define SERVER_STRING "227 Entering Passive Mode (" +#define CLIENT_STRING "PORT " + + +/* + * List of ports (up to IP_VS_APP_MAX_PORTS) to be handled by helper + * First port is set to the default port. + */ +static unsigned short ports[IP_VS_APP_MAX_PORTS] = {21, 0}; +module_param_array(ports, ushort, NULL, 0); +MODULE_PARM_DESC(ports, "Ports to monitor for FTP control commands"); + + +/* Dummy variable */ +static int ip_vs_ftp_pasv; + + +static int +ip_vs_ftp_init_conn(struct ip_vs_app *app, struct ip_vs_conn *cp) +{ + return 0; +} + + +static int +ip_vs_ftp_done_conn(struct ip_vs_app *app, struct ip_vs_conn *cp) +{ + return 0; +} + + +/* + * Get <addr,port> from the string "xxx.xxx.xxx.xxx,ppp,ppp", started + * with the "pattern" and terminated with the "term" character. + * <addr,port> is in network order. + */ +static int ip_vs_ftp_get_addrport(char *data, char *data_limit, + const char *pattern, size_t plen, char term, + __be32 *addr, __be16 *port, + char **start, char **end) +{ + unsigned char p[6]; + int i = 0; + + if (data_limit - data < plen) { + /* check if there is partial match */ + if (strnicmp(data, pattern, data_limit - data) == 0) + return -1; + else + return 0; + } + + if (strnicmp(data, pattern, plen) != 0) { + return 0; + } + *start = data + plen; + + for (data = *start; *data != term; data++) { + if (data == data_limit) + return -1; + } + *end = data; + + memset(p, 0, sizeof(p)); + for (data = *start; data != *end; data++) { + if (*data >= '0' && *data <= '9') { + p[i] = p[i]*10 + *data - '0'; + } else if (*data == ',' && i < 5) { + i++; + } else { + /* unexpected character */ + return -1; + } + } + + if (i != 5) + return -1; + + *addr = get_unaligned((__be32 *)p); + *port = get_unaligned((__be16 *)(p + 4)); + return 1; +} + + +/* + * Look at outgoing ftp packets to catch the response to a PASV command + * from the server (inside-to-outside). + * When we see one, we build a connection entry with the client address, + * client port 0 (unknown at the moment), the server address and the + * server port. Mark the current connection entry as a control channel + * of the new entry. All this work is just to make the data connection + * can be scheduled to the right server later. + * + * The outgoing packet should be something like + * "227 Entering Passive Mode (xxx,xxx,xxx,xxx,ppp,ppp)". + * xxx,xxx,xxx,xxx is the server address, ppp,ppp is the server port number. + */ +static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, + struct sk_buff *skb, int *diff) +{ + struct iphdr *iph; + struct tcphdr *th; + char *data, *data_limit; + char *start, *end; + union nf_inet_addr from; + __be16 port; + struct ip_vs_conn *n_cp; + char buf[24]; /* xxx.xxx.xxx.xxx,ppp,ppp\000 */ + unsigned buf_len; + int ret; + +#ifdef CONFIG_IP_VS_IPV6 + /* This application helper doesn't work with IPv6 yet, + * so turn this into a no-op for IPv6 packets + */ + if (cp->af == AF_INET6) + return 1; +#endif + + *diff = 0; + + /* Only useful for established sessions */ + if (cp->state != IP_VS_TCP_S_ESTABLISHED) + return 1; + + /* Linear packets are much easier to deal with. */ + if (!skb_make_writable(skb, skb->len)) + return 0; + + if (cp->app_data == &ip_vs_ftp_pasv) { + iph = ip_hdr(skb); + th = (struct tcphdr *)&(((char *)iph)[iph->ihl*4]); + data = (char *)th + (th->doff << 2); + data_limit = skb_tail_pointer(skb); + + if (ip_vs_ftp_get_addrport(data, data_limit, + SERVER_STRING, + sizeof(SERVER_STRING)-1, ')', + &from.ip, &port, + &start, &end) != 1) + return 1; + + IP_VS_DBG(7, "PASV response (%u.%u.%u.%u:%d) -> " + "%u.%u.%u.%u:%d detected\n", + NIPQUAD(from.ip), ntohs(port), + NIPQUAD(cp->caddr.ip), 0); + + /* + * Now update or create an connection entry for it + */ + n_cp = ip_vs_conn_out_get(AF_INET, iph->protocol, &from, port, + &cp->caddr, 0); + if (!n_cp) { + n_cp = ip_vs_conn_new(AF_INET, IPPROTO_TCP, + &cp->caddr, 0, + &cp->vaddr, port, + &from, port, + IP_VS_CONN_F_NO_CPORT, + cp->dest); + if (!n_cp) + return 0; + + /* add its controller */ + ip_vs_control_add(n_cp, cp); + } + + /* + * Replace the old passive address with the new one + */ + from.ip = n_cp->vaddr.ip; + port = n_cp->vport; + sprintf(buf, "%d,%d,%d,%d,%d,%d", NIPQUAD(from.ip), + (ntohs(port)>>8)&255, ntohs(port)&255); + buf_len = strlen(buf); + + /* + * Calculate required delta-offset to keep TCP happy + */ + *diff = buf_len - (end-start); + + if (*diff == 0) { + /* simply replace it with new passive address */ + memcpy(start, buf, buf_len); + ret = 1; + } else { + ret = !ip_vs_skb_replace(skb, GFP_ATOMIC, start, + end-start, buf, buf_len); + } + + cp->app_data = NULL; + ip_vs_tcp_conn_listen(n_cp); + ip_vs_conn_put(n_cp); + return ret; + } + return 1; +} + + +/* + * Look at incoming ftp packets to catch the PASV/PORT command + * (outside-to-inside). + * + * The incoming packet having the PORT command should be something like + * "PORT xxx,xxx,xxx,xxx,ppp,ppp\n". + * xxx,xxx,xxx,xxx is the client address, ppp,ppp is the client port number. + * In this case, we create a connection entry using the client address and + * port, so that the active ftp data connection from the server can reach + * the client. + */ +static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, + struct sk_buff *skb, int *diff) +{ + struct iphdr *iph; + struct tcphdr *th; + char *data, *data_start, *data_limit; + char *start, *end; + union nf_inet_addr to; + __be16 port; + struct ip_vs_conn *n_cp; + +#ifdef CONFIG_IP_VS_IPV6 + /* This application helper doesn't work with IPv6 yet, + * so turn this into a no-op for IPv6 packets + */ + if (cp->af == AF_INET6) + return 1; +#endif + + /* no diff required for incoming packets */ + *diff = 0; + + /* Only useful for established sessions */ + if (cp->state != IP_VS_TCP_S_ESTABLISHED) + return 1; + + /* Linear packets are much easier to deal with. */ + if (!skb_make_writable(skb, skb->len)) + return 0; + + /* + * Detecting whether it is passive + */ + iph = ip_hdr(skb); + th = (struct tcphdr *)&(((char *)iph)[iph->ihl*4]); + + /* Since there may be OPTIONS in the TCP packet and the HLEN is + the length of the header in 32-bit multiples, it is accurate + to calculate data address by th+HLEN*4 */ + data = data_start = (char *)th + (th->doff << 2); + data_limit = skb_tail_pointer(skb); + + while (data <= data_limit - 6) { + if (strnicmp(data, "PASV\r\n", 6) == 0) { + /* Passive mode on */ + IP_VS_DBG(7, "got PASV at %td of %td\n", + data - data_start, + data_limit - data_start); + cp->app_data = &ip_vs_ftp_pasv; + return 1; + } + data++; + } + + /* + * To support virtual FTP server, the scenerio is as follows: + * FTP client ----> Load Balancer ----> FTP server + * First detect the port number in the application data, + * then create a new connection entry for the coming data + * connection. + */ + if (ip_vs_ftp_get_addrport(data_start, data_limit, + CLIENT_STRING, sizeof(CLIENT_STRING)-1, + '\r', &to.ip, &port, + &start, &end) != 1) + return 1; + + IP_VS_DBG(7, "PORT %u.%u.%u.%u:%d detected\n", + NIPQUAD(to.ip), ntohs(port)); + + /* Passive mode off */ + cp->app_data = NULL; + + /* + * Now update or create a connection entry for it + */ + IP_VS_DBG(7, "protocol %s %u.%u.%u.%u:%d %u.%u.%u.%u:%d\n", + ip_vs_proto_name(iph->protocol), + NIPQUAD(to.ip), ntohs(port), NIPQUAD(cp->vaddr.ip), 0); + + n_cp = ip_vs_conn_in_get(AF_INET, iph->protocol, + &to, port, + &cp->vaddr, htons(ntohs(cp->vport)-1)); + if (!n_cp) { + n_cp = ip_vs_conn_new(AF_INET, IPPROTO_TCP, + &to, port, + &cp->vaddr, htons(ntohs(cp->vport)-1), + &cp->daddr, htons(ntohs(cp->dport)-1), + 0, + cp->dest); + if (!n_cp) + return 0; + + /* add its controller */ + ip_vs_control_add(n_cp, cp); + } + + /* + * Move tunnel to listen state + */ + ip_vs_tcp_conn_listen(n_cp); + ip_vs_conn_put(n_cp); + + return 1; +} + + +static struct ip_vs_app ip_vs_ftp = { + .name = "ftp", + .type = IP_VS_APP_TYPE_FTP, + .protocol = IPPROTO_TCP, + .module = THIS_MODULE, + .incs_list = LIST_HEAD_INIT(ip_vs_ftp.incs_list), + .init_conn = ip_vs_ftp_init_conn, + .done_conn = ip_vs_ftp_done_conn, + .bind_conn = NULL, + .unbind_conn = NULL, + .pkt_out = ip_vs_ftp_out, + .pkt_in = ip_vs_ftp_in, +}; + + +/* + * ip_vs_ftp initialization + */ +static int __init ip_vs_ftp_init(void) +{ + int i, ret; + struct ip_vs_app *app = &ip_vs_ftp; + + ret = register_ip_vs_app(app); + if (ret) + return ret; + + for (i=0; i<IP_VS_APP_MAX_PORTS; i++) { + if (!ports[i]) + continue; + ret = register_ip_vs_app_inc(app, app->protocol, ports[i]); + if (ret) + break; + IP_VS_INFO("%s: loaded support on port[%d] = %d\n", + app->name, i, ports[i]); + } + + if (ret) + unregister_ip_vs_app(app); + + return ret; +} + + +/* + * ip_vs_ftp finish. + */ +static void __exit ip_vs_ftp_exit(void) +{ + unregister_ip_vs_app(&ip_vs_ftp); +} + + +module_init(ip_vs_ftp_init); +module_exit(ip_vs_ftp_exit); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c new file mode 100644 index 00000000000..6ecef3518ca --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -0,0 +1,555 @@ +/* + * IPVS: Locality-Based Least-Connection scheduling module + * + * Authors: Wensong Zhang <wensong@gnuchina.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * Martin Hamilton : fixed the terrible locking bugs + * *lock(tbl->lock) ==> *lock(&tbl->lock) + * Wensong Zhang : fixed the uninitilized tbl->lock bug + * Wensong Zhang : added doing full expiration check to + * collect stale entries of 24+ hours when + * no partial expire check in a half hour + * Julian Anastasov : replaced del_timer call with del_timer_sync + * to avoid the possible race between timer + * handler and del_timer thread in SMP + * + */ + +/* + * The lblc algorithm is as follows (pseudo code): + * + * if cachenode[dest_ip] is null then + * n, cachenode[dest_ip] <- {weighted least-conn node}; + * else + * n <- cachenode[dest_ip]; + * if (n is dead) OR + * (n.conns>n.weight AND + * there is a node m with m.conns<m.weight/2) then + * n, cachenode[dest_ip] <- {weighted least-conn node}; + * + * return n; + * + * Thanks must go to Wenzhuo Zhang for talking WCCP to me and pushing + * me to write this module. + */ + +#include <linux/ip.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/jiffies.h> + +/* for sysctl */ +#include <linux/fs.h> +#include <linux/sysctl.h> + +#include <net/ip_vs.h> + + +/* + * It is for garbage collection of stale IPVS lblc entries, + * when the table is full. + */ +#define CHECK_EXPIRE_INTERVAL (60*HZ) +#define ENTRY_TIMEOUT (6*60*HZ) + +/* + * It is for full expiration check. + * When there is no partial expiration check (garbage collection) + * in a half hour, do a full expiration check to collect stale + * entries that haven't been touched for a day. + */ +#define COUNT_FOR_FULL_EXPIRATION 30 +static int sysctl_ip_vs_lblc_expiration = 24*60*60*HZ; + + +/* + * for IPVS lblc entry hash table + */ +#ifndef CONFIG_IP_VS_LBLC_TAB_BITS +#define CONFIG_IP_VS_LBLC_TAB_BITS 10 +#endif +#define IP_VS_LBLC_TAB_BITS CONFIG_IP_VS_LBLC_TAB_BITS +#define IP_VS_LBLC_TAB_SIZE (1 << IP_VS_LBLC_TAB_BITS) +#define IP_VS_LBLC_TAB_MASK (IP_VS_LBLC_TAB_SIZE - 1) + + +/* + * IPVS lblc entry represents an association between destination + * IP address and its destination server + */ +struct ip_vs_lblc_entry { + struct list_head list; + __be32 addr; /* destination IP address */ + struct ip_vs_dest *dest; /* real server (cache) */ + unsigned long lastuse; /* last used time */ +}; + + +/* + * IPVS lblc hash table + */ +struct ip_vs_lblc_table { + struct list_head bucket[IP_VS_LBLC_TAB_SIZE]; /* hash bucket */ + atomic_t entries; /* number of entries */ + int max_size; /* maximum size of entries */ + struct timer_list periodic_timer; /* collect stale entries */ + int rover; /* rover for expire check */ + int counter; /* counter for no expire */ +}; + + +/* + * IPVS LBLC sysctl table + */ + +static ctl_table vs_vars_table[] = { + { + .procname = "lblc_expiration", + .data = &sysctl_ip_vs_lblc_expiration, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static struct ctl_table_header * sysctl_header; + +static inline void ip_vs_lblc_free(struct ip_vs_lblc_entry *en) +{ + list_del(&en->list); + /* + * We don't kfree dest because it is refered either by its service + * or the trash dest list. + */ + atomic_dec(&en->dest->refcnt); + kfree(en); +} + + +/* + * Returns hash value for IPVS LBLC entry + */ +static inline unsigned ip_vs_lblc_hashkey(__be32 addr) +{ + return (ntohl(addr)*2654435761UL) & IP_VS_LBLC_TAB_MASK; +} + + +/* + * Hash an entry in the ip_vs_lblc_table. + * returns bool success. + */ +static void +ip_vs_lblc_hash(struct ip_vs_lblc_table *tbl, struct ip_vs_lblc_entry *en) +{ + unsigned hash = ip_vs_lblc_hashkey(en->addr); + + list_add(&en->list, &tbl->bucket[hash]); + atomic_inc(&tbl->entries); +} + + +/* + * Get ip_vs_lblc_entry associated with supplied parameters. Called under read + * lock + */ +static inline struct ip_vs_lblc_entry * +ip_vs_lblc_get(struct ip_vs_lblc_table *tbl, __be32 addr) +{ + unsigned hash = ip_vs_lblc_hashkey(addr); + struct ip_vs_lblc_entry *en; + + list_for_each_entry(en, &tbl->bucket[hash], list) + if (en->addr == addr) + return en; + + return NULL; +} + + +/* + * Create or update an ip_vs_lblc_entry, which is a mapping of a destination IP + * address to a server. Called under write lock. + */ +static inline struct ip_vs_lblc_entry * +ip_vs_lblc_new(struct ip_vs_lblc_table *tbl, __be32 daddr, + struct ip_vs_dest *dest) +{ + struct ip_vs_lblc_entry *en; + + en = ip_vs_lblc_get(tbl, daddr); + if (!en) { + en = kmalloc(sizeof(*en), GFP_ATOMIC); + if (!en) { + IP_VS_ERR("ip_vs_lblc_new(): no memory\n"); + return NULL; + } + + en->addr = daddr; + en->lastuse = jiffies; + + atomic_inc(&dest->refcnt); + en->dest = dest; + + ip_vs_lblc_hash(tbl, en); + } else if (en->dest != dest) { + atomic_dec(&en->dest->refcnt); + atomic_inc(&dest->refcnt); + en->dest = dest; + } + + return en; +} + + +/* + * Flush all the entries of the specified table. + */ +static void ip_vs_lblc_flush(struct ip_vs_lblc_table *tbl) +{ + struct ip_vs_lblc_entry *en, *nxt; + int i; + + for (i=0; i<IP_VS_LBLC_TAB_SIZE; i++) { + list_for_each_entry_safe(en, nxt, &tbl->bucket[i], list) { + ip_vs_lblc_free(en); + atomic_dec(&tbl->entries); + } + } +} + + +static inline void ip_vs_lblc_full_check(struct ip_vs_service *svc) +{ + struct ip_vs_lblc_table *tbl = svc->sched_data; + struct ip_vs_lblc_entry *en, *nxt; + unsigned long now = jiffies; + int i, j; + + for (i=0, j=tbl->rover; i<IP_VS_LBLC_TAB_SIZE; i++) { + j = (j + 1) & IP_VS_LBLC_TAB_MASK; + + write_lock(&svc->sched_lock); + list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { + if (time_before(now, + en->lastuse + sysctl_ip_vs_lblc_expiration)) + continue; + + ip_vs_lblc_free(en); + atomic_dec(&tbl->entries); + } + write_unlock(&svc->sched_lock); + } + tbl->rover = j; +} + + +/* + * Periodical timer handler for IPVS lblc table + * It is used to collect stale entries when the number of entries + * exceeds the maximum size of the table. + * + * Fixme: we probably need more complicated algorithm to collect + * entries that have not been used for a long time even + * if the number of entries doesn't exceed the maximum size + * of the table. + * The full expiration check is for this purpose now. + */ +static void ip_vs_lblc_check_expire(unsigned long data) +{ + struct ip_vs_service *svc = (struct ip_vs_service *) data; + struct ip_vs_lblc_table *tbl = svc->sched_data; + unsigned long now = jiffies; + int goal; + int i, j; + struct ip_vs_lblc_entry *en, *nxt; + + if ((tbl->counter % COUNT_FOR_FULL_EXPIRATION) == 0) { + /* do full expiration check */ + ip_vs_lblc_full_check(svc); + tbl->counter = 1; + goto out; + } + + if (atomic_read(&tbl->entries) <= tbl->max_size) { + tbl->counter++; + goto out; + } + + goal = (atomic_read(&tbl->entries) - tbl->max_size)*4/3; + if (goal > tbl->max_size/2) + goal = tbl->max_size/2; + + for (i=0, j=tbl->rover; i<IP_VS_LBLC_TAB_SIZE; i++) { + j = (j + 1) & IP_VS_LBLC_TAB_MASK; + + write_lock(&svc->sched_lock); + list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { + if (time_before(now, en->lastuse + ENTRY_TIMEOUT)) + continue; + + ip_vs_lblc_free(en); + atomic_dec(&tbl->entries); + goal--; + } + write_unlock(&svc->sched_lock); + if (goal <= 0) + break; + } + tbl->rover = j; + + out: + mod_timer(&tbl->periodic_timer, jiffies+CHECK_EXPIRE_INTERVAL); +} + + +static int ip_vs_lblc_init_svc(struct ip_vs_service *svc) +{ + int i; + struct ip_vs_lblc_table *tbl; + + /* + * Allocate the ip_vs_lblc_table for this service + */ + tbl = kmalloc(sizeof(*tbl), GFP_ATOMIC); + if (tbl == NULL) { + IP_VS_ERR("ip_vs_lblc_init_svc(): no memory\n"); + return -ENOMEM; + } + svc->sched_data = tbl; + IP_VS_DBG(6, "LBLC hash table (memory=%Zdbytes) allocated for " + "current service\n", sizeof(*tbl)); + + /* + * Initialize the hash buckets + */ + for (i=0; i<IP_VS_LBLC_TAB_SIZE; i++) { + INIT_LIST_HEAD(&tbl->bucket[i]); + } + tbl->max_size = IP_VS_LBLC_TAB_SIZE*16; + tbl->rover = 0; + tbl->counter = 1; + + /* + * Hook periodic timer for garbage collection + */ + setup_timer(&tbl->periodic_timer, ip_vs_lblc_check_expire, + (unsigned long)svc); + mod_timer(&tbl->periodic_timer, jiffies + CHECK_EXPIRE_INTERVAL); + + return 0; +} + + +static int ip_vs_lblc_done_svc(struct ip_vs_service *svc) +{ + struct ip_vs_lblc_table *tbl = svc->sched_data; + + /* remove periodic timer */ + del_timer_sync(&tbl->periodic_timer); + + /* got to clean up table entries here */ + ip_vs_lblc_flush(tbl); + + /* release the table itself */ + kfree(tbl); + IP_VS_DBG(6, "LBLC hash table (memory=%Zdbytes) released\n", + sizeof(*tbl)); + + return 0; +} + + +static inline struct ip_vs_dest * +__ip_vs_lblc_schedule(struct ip_vs_service *svc, struct iphdr *iph) +{ + struct ip_vs_dest *dest, *least; + int loh, doh; + + /* + * We think the overhead of processing active connections is fifty + * times higher than that of inactive connections in average. (This + * fifty times might not be accurate, we will change it later.) We + * use the following formula to estimate the overhead: + * dest->activeconns*50 + dest->inactconns + * and the load: + * (dest overhead) / dest->weight + * + * Remember -- no floats in kernel mode!!! + * The comparison of h1*w2 > h2*w1 is equivalent to that of + * h1/w1 > h2/w2 + * if every weight is larger than zero. + * + * The server with weight=0 is quiesced and will not receive any + * new connection. + */ + list_for_each_entry(dest, &svc->destinations, n_list) { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + if (atomic_read(&dest->weight) > 0) { + least = dest; + loh = atomic_read(&least->activeconns) * 50 + + atomic_read(&least->inactconns); + goto nextstage; + } + } + return NULL; + + /* + * Find the destination with the least load. + */ + nextstage: + list_for_each_entry_continue(dest, &svc->destinations, n_list) { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + + doh = atomic_read(&dest->activeconns) * 50 + + atomic_read(&dest->inactconns); + if (loh * atomic_read(&dest->weight) > + doh * atomic_read(&least->weight)) { + least = dest; + loh = doh; + } + } + + IP_VS_DBG(6, "LBLC: server %d.%d.%d.%d:%d " + "activeconns %d refcnt %d weight %d overhead %d\n", + NIPQUAD(least->addr.ip), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->refcnt), + atomic_read(&least->weight), loh); + + return least; +} + + +/* + * If this destination server is overloaded and there is a less loaded + * server, then return true. + */ +static inline int +is_overloaded(struct ip_vs_dest *dest, struct ip_vs_service *svc) +{ + if (atomic_read(&dest->activeconns) > atomic_read(&dest->weight)) { + struct ip_vs_dest *d; + + list_for_each_entry(d, &svc->destinations, n_list) { + if (atomic_read(&d->activeconns)*2 + < atomic_read(&d->weight)) { + return 1; + } + } + } + return 0; +} + + +/* + * Locality-Based (weighted) Least-Connection scheduling + */ +static struct ip_vs_dest * +ip_vs_lblc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_lblc_table *tbl = svc->sched_data; + struct iphdr *iph = ip_hdr(skb); + struct ip_vs_dest *dest = NULL; + struct ip_vs_lblc_entry *en; + + IP_VS_DBG(6, "ip_vs_lblc_schedule(): Scheduling...\n"); + + /* First look in our cache */ + read_lock(&svc->sched_lock); + en = ip_vs_lblc_get(tbl, iph->daddr); + if (en) { + /* We only hold a read lock, but this is atomic */ + en->lastuse = jiffies; + + /* + * If the destination is not available, i.e. it's in the trash, + * we must ignore it, as it may be removed from under our feet, + * if someone drops our reference count. Our caller only makes + * sure that destinations, that are not in the trash, are not + * moved to the trash, while we are scheduling. But anyone can + * free up entries from the trash at any time. + */ + + if (en->dest->flags & IP_VS_DEST_F_AVAILABLE) + dest = en->dest; + } + read_unlock(&svc->sched_lock); + + /* If the destination has a weight and is not overloaded, use it */ + if (dest && atomic_read(&dest->weight) > 0 && !is_overloaded(dest, svc)) + goto out; + + /* No cache entry or it is invalid, time to schedule */ + dest = __ip_vs_lblc_schedule(svc, iph); + if (!dest) { + IP_VS_DBG(1, "no destination available\n"); + return NULL; + } + + /* If we fail to create a cache entry, we'll just use the valid dest */ + write_lock(&svc->sched_lock); + ip_vs_lblc_new(tbl, iph->daddr, dest); + write_unlock(&svc->sched_lock); + +out: + IP_VS_DBG(6, "LBLC: destination IP address %u.%u.%u.%u " + "--> server %u.%u.%u.%u:%d\n", + NIPQUAD(iph->daddr), + NIPQUAD(dest->addr.ip), + ntohs(dest->port)); + + return dest; +} + + +/* + * IPVS LBLC Scheduler structure + */ +static struct ip_vs_scheduler ip_vs_lblc_scheduler = +{ + .name = "lblc", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_lblc_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 0, +#endif + .init_service = ip_vs_lblc_init_svc, + .done_service = ip_vs_lblc_done_svc, + .schedule = ip_vs_lblc_schedule, +}; + + +static int __init ip_vs_lblc_init(void) +{ + int ret; + + sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); + ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler); + if (ret) + unregister_sysctl_table(sysctl_header); + return ret; +} + + +static void __exit ip_vs_lblc_cleanup(void) +{ + unregister_sysctl_table(sysctl_header); + unregister_ip_vs_scheduler(&ip_vs_lblc_scheduler); +} + + +module_init(ip_vs_lblc_init); +module_exit(ip_vs_lblc_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c new file mode 100644 index 00000000000..1f75ea83bcf --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -0,0 +1,755 @@ +/* + * IPVS: Locality-Based Least-Connection with Replication scheduler + * + * Authors: Wensong Zhang <wensong@gnuchina.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * Julian Anastasov : Added the missing (dest->weight>0) + * condition in the ip_vs_dest_set_max. + * + */ + +/* + * The lblc/r algorithm is as follows (pseudo code): + * + * if serverSet[dest_ip] is null then + * n, serverSet[dest_ip] <- {weighted least-conn node}; + * else + * n <- {least-conn (alive) node in serverSet[dest_ip]}; + * if (n is null) OR + * (n.conns>n.weight AND + * there is a node m with m.conns<m.weight/2) then + * n <- {weighted least-conn node}; + * add n to serverSet[dest_ip]; + * if |serverSet[dest_ip]| > 1 AND + * now - serverSet[dest_ip].lastMod > T then + * m <- {most conn node in serverSet[dest_ip]}; + * remove m from serverSet[dest_ip]; + * if serverSet[dest_ip] changed then + * serverSet[dest_ip].lastMod <- now; + * + * return n; + * + */ + +#include <linux/ip.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/jiffies.h> + +/* for sysctl */ +#include <linux/fs.h> +#include <linux/sysctl.h> +#include <net/net_namespace.h> + +#include <net/ip_vs.h> + + +/* + * It is for garbage collection of stale IPVS lblcr entries, + * when the table is full. + */ +#define CHECK_EXPIRE_INTERVAL (60*HZ) +#define ENTRY_TIMEOUT (6*60*HZ) + +/* + * It is for full expiration check. + * When there is no partial expiration check (garbage collection) + * in a half hour, do a full expiration check to collect stale + * entries that haven't been touched for a day. + */ +#define COUNT_FOR_FULL_EXPIRATION 30 +static int sysctl_ip_vs_lblcr_expiration = 24*60*60*HZ; + + +/* + * for IPVS lblcr entry hash table + */ +#ifndef CONFIG_IP_VS_LBLCR_TAB_BITS +#define CONFIG_IP_VS_LBLCR_TAB_BITS 10 +#endif +#define IP_VS_LBLCR_TAB_BITS CONFIG_IP_VS_LBLCR_TAB_BITS +#define IP_VS_LBLCR_TAB_SIZE (1 << IP_VS_LBLCR_TAB_BITS) +#define IP_VS_LBLCR_TAB_MASK (IP_VS_LBLCR_TAB_SIZE - 1) + + +/* + * IPVS destination set structure and operations + */ +struct ip_vs_dest_list { + struct ip_vs_dest_list *next; /* list link */ + struct ip_vs_dest *dest; /* destination server */ +}; + +struct ip_vs_dest_set { + atomic_t size; /* set size */ + unsigned long lastmod; /* last modified time */ + struct ip_vs_dest_list *list; /* destination list */ + rwlock_t lock; /* lock for this list */ +}; + + +static struct ip_vs_dest_list * +ip_vs_dest_set_insert(struct ip_vs_dest_set *set, struct ip_vs_dest *dest) +{ + struct ip_vs_dest_list *e; + + for (e=set->list; e!=NULL; e=e->next) { + if (e->dest == dest) + /* already existed */ + return NULL; + } + + e = kmalloc(sizeof(*e), GFP_ATOMIC); + if (e == NULL) { + IP_VS_ERR("ip_vs_dest_set_insert(): no memory\n"); + return NULL; + } + + atomic_inc(&dest->refcnt); + e->dest = dest; + + /* link it to the list */ + e->next = set->list; + set->list = e; + atomic_inc(&set->size); + + set->lastmod = jiffies; + return e; +} + +static void +ip_vs_dest_set_erase(struct ip_vs_dest_set *set, struct ip_vs_dest *dest) +{ + struct ip_vs_dest_list *e, **ep; + + for (ep=&set->list, e=*ep; e!=NULL; e=*ep) { + if (e->dest == dest) { + /* HIT */ + *ep = e->next; + atomic_dec(&set->size); + set->lastmod = jiffies; + atomic_dec(&e->dest->refcnt); + kfree(e); + break; + } + ep = &e->next; + } +} + +static void ip_vs_dest_set_eraseall(struct ip_vs_dest_set *set) +{ + struct ip_vs_dest_list *e, **ep; + + write_lock(&set->lock); + for (ep=&set->list, e=*ep; e!=NULL; e=*ep) { + *ep = e->next; + /* + * We don't kfree dest because it is refered either + * by its service or by the trash dest list. + */ + atomic_dec(&e->dest->refcnt); + kfree(e); + } + write_unlock(&set->lock); +} + +/* get weighted least-connection node in the destination set */ +static inline struct ip_vs_dest *ip_vs_dest_set_min(struct ip_vs_dest_set *set) +{ + register struct ip_vs_dest_list *e; + struct ip_vs_dest *dest, *least; + int loh, doh; + + if (set == NULL) + return NULL; + + /* select the first destination server, whose weight > 0 */ + for (e=set->list; e!=NULL; e=e->next) { + least = e->dest; + if (least->flags & IP_VS_DEST_F_OVERLOAD) + continue; + + if ((atomic_read(&least->weight) > 0) + && (least->flags & IP_VS_DEST_F_AVAILABLE)) { + loh = atomic_read(&least->activeconns) * 50 + + atomic_read(&least->inactconns); + goto nextstage; + } + } + return NULL; + + /* find the destination with the weighted least load */ + nextstage: + for (e=e->next; e!=NULL; e=e->next) { + dest = e->dest; + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + + doh = atomic_read(&dest->activeconns) * 50 + + atomic_read(&dest->inactconns); + if ((loh * atomic_read(&dest->weight) > + doh * atomic_read(&least->weight)) + && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { + least = dest; + loh = doh; + } + } + + IP_VS_DBG(6, "ip_vs_dest_set_min: server %d.%d.%d.%d:%d " + "activeconns %d refcnt %d weight %d overhead %d\n", + NIPQUAD(least->addr.ip), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->refcnt), + atomic_read(&least->weight), loh); + return least; +} + + +/* get weighted most-connection node in the destination set */ +static inline struct ip_vs_dest *ip_vs_dest_set_max(struct ip_vs_dest_set *set) +{ + register struct ip_vs_dest_list *e; + struct ip_vs_dest *dest, *most; + int moh, doh; + + if (set == NULL) + return NULL; + + /* select the first destination server, whose weight > 0 */ + for (e=set->list; e!=NULL; e=e->next) { + most = e->dest; + if (atomic_read(&most->weight) > 0) { + moh = atomic_read(&most->activeconns) * 50 + + atomic_read(&most->inactconns); + goto nextstage; + } + } + return NULL; + + /* find the destination with the weighted most load */ + nextstage: + for (e=e->next; e!=NULL; e=e->next) { + dest = e->dest; + doh = atomic_read(&dest->activeconns) * 50 + + atomic_read(&dest->inactconns); + /* moh/mw < doh/dw ==> moh*dw < doh*mw, where mw,dw>0 */ + if ((moh * atomic_read(&dest->weight) < + doh * atomic_read(&most->weight)) + && (atomic_read(&dest->weight) > 0)) { + most = dest; + moh = doh; + } + } + + IP_VS_DBG(6, "ip_vs_dest_set_max: server %d.%d.%d.%d:%d " + "activeconns %d refcnt %d weight %d overhead %d\n", + NIPQUAD(most->addr.ip), ntohs(most->port), + atomic_read(&most->activeconns), + atomic_read(&most->refcnt), + atomic_read(&most->weight), moh); + return most; +} + + +/* + * IPVS lblcr entry represents an association between destination + * IP address and its destination server set + */ +struct ip_vs_lblcr_entry { + struct list_head list; + __be32 addr; /* destination IP address */ + struct ip_vs_dest_set set; /* destination server set */ + unsigned long lastuse; /* last used time */ +}; + + +/* + * IPVS lblcr hash table + */ +struct ip_vs_lblcr_table { + struct list_head bucket[IP_VS_LBLCR_TAB_SIZE]; /* hash bucket */ + atomic_t entries; /* number of entries */ + int max_size; /* maximum size of entries */ + struct timer_list periodic_timer; /* collect stale entries */ + int rover; /* rover for expire check */ + int counter; /* counter for no expire */ +}; + + +/* + * IPVS LBLCR sysctl table + */ + +static ctl_table vs_vars_table[] = { + { + .procname = "lblcr_expiration", + .data = &sysctl_ip_vs_lblcr_expiration, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static struct ctl_table_header * sysctl_header; + +static inline void ip_vs_lblcr_free(struct ip_vs_lblcr_entry *en) +{ + list_del(&en->list); + ip_vs_dest_set_eraseall(&en->set); + kfree(en); +} + + +/* + * Returns hash value for IPVS LBLCR entry + */ +static inline unsigned ip_vs_lblcr_hashkey(__be32 addr) +{ + return (ntohl(addr)*2654435761UL) & IP_VS_LBLCR_TAB_MASK; +} + + +/* + * Hash an entry in the ip_vs_lblcr_table. + * returns bool success. + */ +static void +ip_vs_lblcr_hash(struct ip_vs_lblcr_table *tbl, struct ip_vs_lblcr_entry *en) +{ + unsigned hash = ip_vs_lblcr_hashkey(en->addr); + + list_add(&en->list, &tbl->bucket[hash]); + atomic_inc(&tbl->entries); +} + + +/* + * Get ip_vs_lblcr_entry associated with supplied parameters. Called under + * read lock. + */ +static inline struct ip_vs_lblcr_entry * +ip_vs_lblcr_get(struct ip_vs_lblcr_table *tbl, __be32 addr) +{ + unsigned hash = ip_vs_lblcr_hashkey(addr); + struct ip_vs_lblcr_entry *en; + + list_for_each_entry(en, &tbl->bucket[hash], list) + if (en->addr == addr) + return en; + + return NULL; +} + + +/* + * Create or update an ip_vs_lblcr_entry, which is a mapping of a destination + * IP address to a server. Called under write lock. + */ +static inline struct ip_vs_lblcr_entry * +ip_vs_lblcr_new(struct ip_vs_lblcr_table *tbl, __be32 daddr, + struct ip_vs_dest *dest) +{ + struct ip_vs_lblcr_entry *en; + + en = ip_vs_lblcr_get(tbl, daddr); + if (!en) { + en = kmalloc(sizeof(*en), GFP_ATOMIC); + if (!en) { + IP_VS_ERR("ip_vs_lblcr_new(): no memory\n"); + return NULL; + } + + en->addr = daddr; + en->lastuse = jiffies; + + /* initilize its dest set */ + atomic_set(&(en->set.size), 0); + en->set.list = NULL; + rwlock_init(&en->set.lock); + + ip_vs_lblcr_hash(tbl, en); + } + + write_lock(&en->set.lock); + ip_vs_dest_set_insert(&en->set, dest); + write_unlock(&en->set.lock); + + return en; +} + + +/* + * Flush all the entries of the specified table. + */ +static void ip_vs_lblcr_flush(struct ip_vs_lblcr_table *tbl) +{ + int i; + struct ip_vs_lblcr_entry *en, *nxt; + + /* No locking required, only called during cleanup. */ + for (i=0; i<IP_VS_LBLCR_TAB_SIZE; i++) { + list_for_each_entry_safe(en, nxt, &tbl->bucket[i], list) { + ip_vs_lblcr_free(en); + } + } +} + + +static inline void ip_vs_lblcr_full_check(struct ip_vs_service *svc) +{ + struct ip_vs_lblcr_table *tbl = svc->sched_data; + unsigned long now = jiffies; + int i, j; + struct ip_vs_lblcr_entry *en, *nxt; + + for (i=0, j=tbl->rover; i<IP_VS_LBLCR_TAB_SIZE; i++) { + j = (j + 1) & IP_VS_LBLCR_TAB_MASK; + + write_lock(&svc->sched_lock); + list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { + if (time_after(en->lastuse+sysctl_ip_vs_lblcr_expiration, + now)) + continue; + + ip_vs_lblcr_free(en); + atomic_dec(&tbl->entries); + } + write_unlock(&svc->sched_lock); + } + tbl->rover = j; +} + + +/* + * Periodical timer handler for IPVS lblcr table + * It is used to collect stale entries when the number of entries + * exceeds the maximum size of the table. + * + * Fixme: we probably need more complicated algorithm to collect + * entries that have not been used for a long time even + * if the number of entries doesn't exceed the maximum size + * of the table. + * The full expiration check is for this purpose now. + */ +static void ip_vs_lblcr_check_expire(unsigned long data) +{ + struct ip_vs_service *svc = (struct ip_vs_service *) data; + struct ip_vs_lblcr_table *tbl = svc->sched_data; + unsigned long now = jiffies; + int goal; + int i, j; + struct ip_vs_lblcr_entry *en, *nxt; + + if ((tbl->counter % COUNT_FOR_FULL_EXPIRATION) == 0) { + /* do full expiration check */ + ip_vs_lblcr_full_check(svc); + tbl->counter = 1; + goto out; + } + + if (atomic_read(&tbl->entries) <= tbl->max_size) { + tbl->counter++; + goto out; + } + + goal = (atomic_read(&tbl->entries) - tbl->max_size)*4/3; + if (goal > tbl->max_size/2) + goal = tbl->max_size/2; + + for (i=0, j=tbl->rover; i<IP_VS_LBLCR_TAB_SIZE; i++) { + j = (j + 1) & IP_VS_LBLCR_TAB_MASK; + + write_lock(&svc->sched_lock); + list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { + if (time_before(now, en->lastuse+ENTRY_TIMEOUT)) + continue; + + ip_vs_lblcr_free(en); + atomic_dec(&tbl->entries); + goal--; + } + write_unlock(&svc->sched_lock); + if (goal <= 0) + break; + } + tbl->rover = j; + + out: + mod_timer(&tbl->periodic_timer, jiffies+CHECK_EXPIRE_INTERVAL); +} + +static int ip_vs_lblcr_init_svc(struct ip_vs_service *svc) +{ + int i; + struct ip_vs_lblcr_table *tbl; + + /* + * Allocate the ip_vs_lblcr_table for this service + */ + tbl = kmalloc(sizeof(*tbl), GFP_ATOMIC); + if (tbl == NULL) { + IP_VS_ERR("ip_vs_lblcr_init_svc(): no memory\n"); + return -ENOMEM; + } + svc->sched_data = tbl; + IP_VS_DBG(6, "LBLCR hash table (memory=%Zdbytes) allocated for " + "current service\n", sizeof(*tbl)); + + /* + * Initialize the hash buckets + */ + for (i=0; i<IP_VS_LBLCR_TAB_SIZE; i++) { + INIT_LIST_HEAD(&tbl->bucket[i]); + } + tbl->max_size = IP_VS_LBLCR_TAB_SIZE*16; + tbl->rover = 0; + tbl->counter = 1; + + /* + * Hook periodic timer for garbage collection + */ + setup_timer(&tbl->periodic_timer, ip_vs_lblcr_check_expire, + (unsigned long)svc); + mod_timer(&tbl->periodic_timer, jiffies + CHECK_EXPIRE_INTERVAL); + + return 0; +} + + +static int ip_vs_lblcr_done_svc(struct ip_vs_service *svc) +{ + struct ip_vs_lblcr_table *tbl = svc->sched_data; + + /* remove periodic timer */ + del_timer_sync(&tbl->periodic_timer); + + /* got to clean up table entries here */ + ip_vs_lblcr_flush(tbl); + + /* release the table itself */ + kfree(tbl); + IP_VS_DBG(6, "LBLCR hash table (memory=%Zdbytes) released\n", + sizeof(*tbl)); + + return 0; +} + + +static inline struct ip_vs_dest * +__ip_vs_lblcr_schedule(struct ip_vs_service *svc, struct iphdr *iph) +{ + struct ip_vs_dest *dest, *least; + int loh, doh; + + /* + * We think the overhead of processing active connections is fifty + * times higher than that of inactive connections in average. (This + * fifty times might not be accurate, we will change it later.) We + * use the following formula to estimate the overhead: + * dest->activeconns*50 + dest->inactconns + * and the load: + * (dest overhead) / dest->weight + * + * Remember -- no floats in kernel mode!!! + * The comparison of h1*w2 > h2*w1 is equivalent to that of + * h1/w1 > h2/w2 + * if every weight is larger than zero. + * + * The server with weight=0 is quiesced and will not receive any + * new connection. + */ + list_for_each_entry(dest, &svc->destinations, n_list) { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + + if (atomic_read(&dest->weight) > 0) { + least = dest; + loh = atomic_read(&least->activeconns) * 50 + + atomic_read(&least->inactconns); + goto nextstage; + } + } + return NULL; + + /* + * Find the destination with the least load. + */ + nextstage: + list_for_each_entry_continue(dest, &svc->destinations, n_list) { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + + doh = atomic_read(&dest->activeconns) * 50 + + atomic_read(&dest->inactconns); + if (loh * atomic_read(&dest->weight) > + doh * atomic_read(&least->weight)) { + least = dest; + loh = doh; + } + } + + IP_VS_DBG(6, "LBLCR: server %d.%d.%d.%d:%d " + "activeconns %d refcnt %d weight %d overhead %d\n", + NIPQUAD(least->addr.ip), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->refcnt), + atomic_read(&least->weight), loh); + + return least; +} + + +/* + * If this destination server is overloaded and there is a less loaded + * server, then return true. + */ +static inline int +is_overloaded(struct ip_vs_dest *dest, struct ip_vs_service *svc) +{ + if (atomic_read(&dest->activeconns) > atomic_read(&dest->weight)) { + struct ip_vs_dest *d; + + list_for_each_entry(d, &svc->destinations, n_list) { + if (atomic_read(&d->activeconns)*2 + < atomic_read(&d->weight)) { + return 1; + } + } + } + return 0; +} + + +/* + * Locality-Based (weighted) Least-Connection scheduling + */ +static struct ip_vs_dest * +ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_lblcr_table *tbl = svc->sched_data; + struct iphdr *iph = ip_hdr(skb); + struct ip_vs_dest *dest = NULL; + struct ip_vs_lblcr_entry *en; + + IP_VS_DBG(6, "ip_vs_lblcr_schedule(): Scheduling...\n"); + + /* First look in our cache */ + read_lock(&svc->sched_lock); + en = ip_vs_lblcr_get(tbl, iph->daddr); + if (en) { + /* We only hold a read lock, but this is atomic */ + en->lastuse = jiffies; + + /* Get the least loaded destination */ + read_lock(&en->set.lock); + dest = ip_vs_dest_set_min(&en->set); + read_unlock(&en->set.lock); + + /* More than one destination + enough time passed by, cleanup */ + if (atomic_read(&en->set.size) > 1 && + time_after(jiffies, en->set.lastmod + + sysctl_ip_vs_lblcr_expiration)) { + struct ip_vs_dest *m; + + write_lock(&en->set.lock); + m = ip_vs_dest_set_max(&en->set); + if (m) + ip_vs_dest_set_erase(&en->set, m); + write_unlock(&en->set.lock); + } + + /* If the destination is not overloaded, use it */ + if (dest && !is_overloaded(dest, svc)) { + read_unlock(&svc->sched_lock); + goto out; + } + + /* The cache entry is invalid, time to schedule */ + dest = __ip_vs_lblcr_schedule(svc, iph); + if (!dest) { + IP_VS_DBG(1, "no destination available\n"); + read_unlock(&svc->sched_lock); + return NULL; + } + + /* Update our cache entry */ + write_lock(&en->set.lock); + ip_vs_dest_set_insert(&en->set, dest); + write_unlock(&en->set.lock); + } + read_unlock(&svc->sched_lock); + + if (dest) + goto out; + + /* No cache entry, time to schedule */ + dest = __ip_vs_lblcr_schedule(svc, iph); + if (!dest) { + IP_VS_DBG(1, "no destination available\n"); + return NULL; + } + + /* If we fail to create a cache entry, we'll just use the valid dest */ + write_lock(&svc->sched_lock); + ip_vs_lblcr_new(tbl, iph->daddr, dest); + write_unlock(&svc->sched_lock); + +out: + IP_VS_DBG(6, "LBLCR: destination IP address %u.%u.%u.%u " + "--> server %u.%u.%u.%u:%d\n", + NIPQUAD(iph->daddr), + NIPQUAD(dest->addr.ip), + ntohs(dest->port)); + + return dest; +} + + +/* + * IPVS LBLCR Scheduler structure + */ +static struct ip_vs_scheduler ip_vs_lblcr_scheduler = +{ + .name = "lblcr", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_lblcr_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 0, +#endif + .init_service = ip_vs_lblcr_init_svc, + .done_service = ip_vs_lblcr_done_svc, + .schedule = ip_vs_lblcr_schedule, +}; + + +static int __init ip_vs_lblcr_init(void) +{ + int ret; + + sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); + ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler); + if (ret) + unregister_sysctl_table(sysctl_header); + return ret; +} + + +static void __exit ip_vs_lblcr_cleanup(void) +{ + unregister_sysctl_table(sysctl_header); + unregister_ip_vs_scheduler(&ip_vs_lblcr_scheduler); +} + + +module_init(ip_vs_lblcr_init); +module_exit(ip_vs_lblcr_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_lc.c b/net/netfilter/ipvs/ip_vs_lc.c new file mode 100644 index 00000000000..b69f808ac46 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_lc.c @@ -0,0 +1,103 @@ +/* + * IPVS: Least-Connection Scheduling module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * Wensong Zhang : added the ip_vs_lc_update_svc + * Wensong Zhang : added any dest with weight=0 is quiesced + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> + +#include <net/ip_vs.h> + + +static inline unsigned int +ip_vs_lc_dest_overhead(struct ip_vs_dest *dest) +{ + /* + * We think the overhead of processing active connections is 256 + * times higher than that of inactive connections in average. (This + * 256 times might not be accurate, we will change it later) We + * use the following formula to estimate the overhead now: + * dest->activeconns*256 + dest->inactconns + */ + return (atomic_read(&dest->activeconns) << 8) + + atomic_read(&dest->inactconns); +} + + +/* + * Least Connection scheduling + */ +static struct ip_vs_dest * +ip_vs_lc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest, *least = NULL; + unsigned int loh = 0, doh; + + IP_VS_DBG(6, "ip_vs_lc_schedule(): Scheduling...\n"); + + /* + * Simply select the server with the least number of + * (activeconns<<5) + inactconns + * Except whose weight is equal to zero. + * If the weight is equal to zero, it means that the server is + * quiesced, the existing connections to the server still get + * served, but no new connection is assigned to the server. + */ + + list_for_each_entry(dest, &svc->destinations, n_list) { + if ((dest->flags & IP_VS_DEST_F_OVERLOAD) || + atomic_read(&dest->weight) == 0) + continue; + doh = ip_vs_lc_dest_overhead(dest); + if (!least || doh < loh) { + least = dest; + loh = doh; + } + } + + if (least) + IP_VS_DBG_BUF(6, "LC: server %s:%u activeconns %d inactconns %d\n", + IP_VS_DBG_ADDR(svc->af, &least->addr), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->inactconns)); + + return least; +} + + +static struct ip_vs_scheduler ip_vs_lc_scheduler = { + .name = "lc", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_lc_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 1, +#endif + .schedule = ip_vs_lc_schedule, +}; + + +static int __init ip_vs_lc_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_lc_scheduler) ; +} + +static void __exit ip_vs_lc_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_lc_scheduler); +} + +module_init(ip_vs_lc_init); +module_exit(ip_vs_lc_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_nq.c b/net/netfilter/ipvs/ip_vs_nq.c new file mode 100644 index 00000000000..9a2d8033f08 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_nq.c @@ -0,0 +1,138 @@ +/* + * IPVS: Never Queue scheduling module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +/* + * The NQ algorithm adopts a two-speed model. When there is an idle server + * available, the job will be sent to the idle server, instead of waiting + * for a fast one. When there is no idle server available, the job will be + * sent to the server that minimize its expected delay (The Shortest + * Expected Delay scheduling algorithm). + * + * See the following paper for more information: + * A. Weinrib and S. Shenker, Greed is not enough: Adaptive load sharing + * in large heterogeneous systems. In Proceedings IEEE INFOCOM'88, + * pages 986-994, 1988. + * + * Thanks must go to Marko Buuri <marko@buuri.name> for talking NQ to me. + * + * The difference between NQ and SED is that NQ can improve overall + * system utilization. + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> + +#include <net/ip_vs.h> + + +static inline unsigned int +ip_vs_nq_dest_overhead(struct ip_vs_dest *dest) +{ + /* + * We only use the active connection number in the cost + * calculation here. + */ + return atomic_read(&dest->activeconns) + 1; +} + + +/* + * Weighted Least Connection scheduling + */ +static struct ip_vs_dest * +ip_vs_nq_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest, *least = NULL; + unsigned int loh = 0, doh; + + IP_VS_DBG(6, "ip_vs_nq_schedule(): Scheduling...\n"); + + /* + * We calculate the load of each dest server as follows: + * (server expected overhead) / dest->weight + * + * Remember -- no floats in kernel mode!!! + * The comparison of h1*w2 > h2*w1 is equivalent to that of + * h1/w1 > h2/w2 + * if every weight is larger than zero. + * + * The server with weight=0 is quiesced and will not receive any + * new connections. + */ + + list_for_each_entry(dest, &svc->destinations, n_list) { + + if (dest->flags & IP_VS_DEST_F_OVERLOAD || + !atomic_read(&dest->weight)) + continue; + + doh = ip_vs_nq_dest_overhead(dest); + + /* return the server directly if it is idle */ + if (atomic_read(&dest->activeconns) == 0) { + least = dest; + loh = doh; + goto out; + } + + if (!least || + (loh * atomic_read(&dest->weight) > + doh * atomic_read(&least->weight))) { + least = dest; + loh = doh; + } + } + + if (!least) + return NULL; + + out: + IP_VS_DBG_BUF(6, "NQ: server %s:%u " + "activeconns %d refcnt %d weight %d overhead %d\n", + IP_VS_DBG_ADDR(svc->af, &least->addr), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->refcnt), + atomic_read(&least->weight), loh); + + return least; +} + + +static struct ip_vs_scheduler ip_vs_nq_scheduler = +{ + .name = "nq", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_nq_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 1, +#endif + .schedule = ip_vs_nq_schedule, +}; + + +static int __init ip_vs_nq_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_nq_scheduler); +} + +static void __exit ip_vs_nq_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_nq_scheduler); +} + +module_init(ip_vs_nq_init); +module_exit(ip_vs_nq_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c new file mode 100644 index 00000000000..0791f9e08fe --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -0,0 +1,288 @@ +/* + * ip_vs_proto.c: transport protocol load balancing support for IPVS + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/ip.h> +#include <net/protocol.h> +#include <net/tcp.h> +#include <net/udp.h> +#include <asm/system.h> +#include <linux/stat.h> +#include <linux/proc_fs.h> + +#include <net/ip_vs.h> + + +/* + * IPVS protocols can only be registered/unregistered when the ipvs + * module is loaded/unloaded, so no lock is needed in accessing the + * ipvs protocol table. + */ + +#define IP_VS_PROTO_TAB_SIZE 32 /* must be power of 2 */ +#define IP_VS_PROTO_HASH(proto) ((proto) & (IP_VS_PROTO_TAB_SIZE-1)) + +static struct ip_vs_protocol *ip_vs_proto_table[IP_VS_PROTO_TAB_SIZE]; + + +/* + * register an ipvs protocol + */ +static int __used __init register_ip_vs_protocol(struct ip_vs_protocol *pp) +{ + unsigned hash = IP_VS_PROTO_HASH(pp->protocol); + + pp->next = ip_vs_proto_table[hash]; + ip_vs_proto_table[hash] = pp; + + if (pp->init != NULL) + pp->init(pp); + + return 0; +} + + +/* + * unregister an ipvs protocol + */ +static int unregister_ip_vs_protocol(struct ip_vs_protocol *pp) +{ + struct ip_vs_protocol **pp_p; + unsigned hash = IP_VS_PROTO_HASH(pp->protocol); + + pp_p = &ip_vs_proto_table[hash]; + for (; *pp_p; pp_p = &(*pp_p)->next) { + if (*pp_p == pp) { + *pp_p = pp->next; + if (pp->exit != NULL) + pp->exit(pp); + return 0; + } + } + + return -ESRCH; +} + + +/* + * get ip_vs_protocol object by its proto. + */ +struct ip_vs_protocol * ip_vs_proto_get(unsigned short proto) +{ + struct ip_vs_protocol *pp; + unsigned hash = IP_VS_PROTO_HASH(proto); + + for (pp = ip_vs_proto_table[hash]; pp; pp = pp->next) { + if (pp->protocol == proto) + return pp; + } + + return NULL; +} + + +/* + * Propagate event for state change to all protocols + */ +void ip_vs_protocol_timeout_change(int flags) +{ + struct ip_vs_protocol *pp; + int i; + + for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) { + for (pp = ip_vs_proto_table[i]; pp; pp = pp->next) { + if (pp->timeout_change) + pp->timeout_change(pp, flags); + } + } +} + + +int * +ip_vs_create_timeout_table(int *table, int size) +{ + return kmemdup(table, size, GFP_ATOMIC); +} + + +/* + * Set timeout value for state specified by name + */ +int +ip_vs_set_state_timeout(int *table, int num, char **names, char *name, int to) +{ + int i; + + if (!table || !name || !to) + return -EINVAL; + + for (i = 0; i < num; i++) { + if (strcmp(names[i], name)) + continue; + table[i] = to * HZ; + return 0; + } + return -ENOENT; +} + + +const char * ip_vs_state_name(__u16 proto, int state) +{ + struct ip_vs_protocol *pp = ip_vs_proto_get(proto); + + if (pp == NULL || pp->state_name == NULL) + return (IPPROTO_IP == proto) ? "NONE" : "ERR!"; + return pp->state_name(state); +} + + +static void +ip_vs_tcpudp_debug_packet_v4(struct ip_vs_protocol *pp, + const struct sk_buff *skb, + int offset, + const char *msg) +{ + char buf[128]; + struct iphdr _iph, *ih; + + ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); + if (ih == NULL) + sprintf(buf, "%s TRUNCATED", pp->name); + else if (ih->frag_off & htons(IP_OFFSET)) + sprintf(buf, "%s %u.%u.%u.%u->%u.%u.%u.%u frag", + pp->name, NIPQUAD(ih->saddr), + NIPQUAD(ih->daddr)); + else { + __be16 _ports[2], *pptr +; + pptr = skb_header_pointer(skb, offset + ih->ihl*4, + sizeof(_ports), _ports); + if (pptr == NULL) + sprintf(buf, "%s TRUNCATED %u.%u.%u.%u->%u.%u.%u.%u", + pp->name, + NIPQUAD(ih->saddr), + NIPQUAD(ih->daddr)); + else + sprintf(buf, "%s %u.%u.%u.%u:%u->%u.%u.%u.%u:%u", + pp->name, + NIPQUAD(ih->saddr), + ntohs(pptr[0]), + NIPQUAD(ih->daddr), + ntohs(pptr[1])); + } + + printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf); +} + +#ifdef CONFIG_IP_VS_IPV6 +static void +ip_vs_tcpudp_debug_packet_v6(struct ip_vs_protocol *pp, + const struct sk_buff *skb, + int offset, + const char *msg) +{ + char buf[192]; + struct ipv6hdr _iph, *ih; + + ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); + if (ih == NULL) + sprintf(buf, "%s TRUNCATED", pp->name); + else if (ih->nexthdr == IPPROTO_FRAGMENT) + sprintf(buf, "%s " NIP6_FMT "->" NIP6_FMT " frag", + pp->name, NIP6(ih->saddr), + NIP6(ih->daddr)); + else { + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, offset + sizeof(struct ipv6hdr), + sizeof(_ports), _ports); + if (pptr == NULL) + sprintf(buf, "%s TRUNCATED " NIP6_FMT "->" NIP6_FMT, + pp->name, + NIP6(ih->saddr), + NIP6(ih->daddr)); + else + sprintf(buf, "%s " NIP6_FMT ":%u->" NIP6_FMT ":%u", + pp->name, + NIP6(ih->saddr), + ntohs(pptr[0]), + NIP6(ih->daddr), + ntohs(pptr[1])); + } + + printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf); +} +#endif + + +void +ip_vs_tcpudp_debug_packet(struct ip_vs_protocol *pp, + const struct sk_buff *skb, + int offset, + const char *msg) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (skb->protocol == htons(ETH_P_IPV6)) + ip_vs_tcpudp_debug_packet_v6(pp, skb, offset, msg); + else +#endif + ip_vs_tcpudp_debug_packet_v4(pp, skb, offset, msg); +} + + +int __init ip_vs_protocol_init(void) +{ + char protocols[64]; +#define REGISTER_PROTOCOL(p) \ + do { \ + register_ip_vs_protocol(p); \ + strcat(protocols, ", "); \ + strcat(protocols, (p)->name); \ + } while (0) + + protocols[0] = '\0'; + protocols[2] = '\0'; +#ifdef CONFIG_IP_VS_PROTO_TCP + REGISTER_PROTOCOL(&ip_vs_protocol_tcp); +#endif +#ifdef CONFIG_IP_VS_PROTO_UDP + REGISTER_PROTOCOL(&ip_vs_protocol_udp); +#endif +#ifdef CONFIG_IP_VS_PROTO_AH + REGISTER_PROTOCOL(&ip_vs_protocol_ah); +#endif +#ifdef CONFIG_IP_VS_PROTO_ESP + REGISTER_PROTOCOL(&ip_vs_protocol_esp); +#endif + IP_VS_INFO("Registered protocols (%s)\n", &protocols[2]); + + return 0; +} + + +void ip_vs_protocol_cleanup(void) +{ + struct ip_vs_protocol *pp; + int i; + + /* unregister all the ipvs protocols */ + for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) { + while ((pp = ip_vs_proto_table[i]) != NULL) + unregister_ip_vs_protocol(pp); + } +} diff --git a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c new file mode 100644 index 00000000000..80ab0c8e5b4 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c @@ -0,0 +1,235 @@ +/* + * ip_vs_proto_ah_esp.c: AH/ESP IPSec load balancing support for IPVS + * + * Authors: Julian Anastasov <ja@ssi.bg>, February 2002 + * Wensong Zhang <wensong@linuxvirtualserver.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/in.h> +#include <linux/ip.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> + +#include <net/ip_vs.h> + + +/* TODO: + +struct isakmp_hdr { + __u8 icookie[8]; + __u8 rcookie[8]; + __u8 np; + __u8 version; + __u8 xchgtype; + __u8 flags; + __u32 msgid; + __u32 length; +}; + +*/ + +#define PORT_ISAKMP 500 + + +static struct ip_vs_conn * +ah_esp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, unsigned int proto_off, + int inverse) +{ + struct ip_vs_conn *cp; + + if (likely(!inverse)) { + cp = ip_vs_conn_in_get(af, IPPROTO_UDP, + &iph->saddr, + htons(PORT_ISAKMP), + &iph->daddr, + htons(PORT_ISAKMP)); + } else { + cp = ip_vs_conn_in_get(af, IPPROTO_UDP, + &iph->daddr, + htons(PORT_ISAKMP), + &iph->saddr, + htons(PORT_ISAKMP)); + } + + if (!cp) { + /* + * We are not sure if the packet is from our + * service, so our conn_schedule hook should return NF_ACCEPT + */ + IP_VS_DBG_BUF(12, "Unknown ISAKMP entry for outin packet " + "%s%s %s->%s\n", + inverse ? "ICMP+" : "", + pp->name, + IP_VS_DBG_ADDR(af, &iph->saddr), + IP_VS_DBG_ADDR(af, &iph->daddr)); + } + + return cp; +} + + +static struct ip_vs_conn * +ah_esp_conn_out_get(int af, const struct sk_buff *skb, + struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, + unsigned int proto_off, + int inverse) +{ + struct ip_vs_conn *cp; + + if (likely(!inverse)) { + cp = ip_vs_conn_out_get(af, IPPROTO_UDP, + &iph->saddr, + htons(PORT_ISAKMP), + &iph->daddr, + htons(PORT_ISAKMP)); + } else { + cp = ip_vs_conn_out_get(af, IPPROTO_UDP, + &iph->daddr, + htons(PORT_ISAKMP), + &iph->saddr, + htons(PORT_ISAKMP)); + } + + if (!cp) { + IP_VS_DBG_BUF(12, "Unknown ISAKMP entry for inout packet " + "%s%s %s->%s\n", + inverse ? "ICMP+" : "", + pp->name, + IP_VS_DBG_ADDR(af, &iph->saddr), + IP_VS_DBG_ADDR(af, &iph->daddr)); + } + + return cp; +} + + +static int +ah_esp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, + int *verdict, struct ip_vs_conn **cpp) +{ + /* + * AH/ESP is only related traffic. Pass the packet to IP stack. + */ + *verdict = NF_ACCEPT; + return 0; +} + + +static void +ah_esp_debug_packet_v4(struct ip_vs_protocol *pp, const struct sk_buff *skb, + int offset, const char *msg) +{ + char buf[256]; + struct iphdr _iph, *ih; + + ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); + if (ih == NULL) + sprintf(buf, "%s TRUNCATED", pp->name); + else + sprintf(buf, "%s %u.%u.%u.%u->%u.%u.%u.%u", + pp->name, NIPQUAD(ih->saddr), + NIPQUAD(ih->daddr)); + + printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf); +} + +#ifdef CONFIG_IP_VS_IPV6 +static void +ah_esp_debug_packet_v6(struct ip_vs_protocol *pp, const struct sk_buff *skb, + int offset, const char *msg) +{ + char buf[256]; + struct ipv6hdr _iph, *ih; + + ih = skb_header_pointer(skb, offset, sizeof(_iph), &_iph); + if (ih == NULL) + sprintf(buf, "%s TRUNCATED", pp->name); + else + sprintf(buf, "%s " NIP6_FMT "->" NIP6_FMT, + pp->name, NIP6(ih->saddr), + NIP6(ih->daddr)); + + printk(KERN_DEBUG "IPVS: %s: %s\n", msg, buf); +} +#endif + +static void +ah_esp_debug_packet(struct ip_vs_protocol *pp, const struct sk_buff *skb, + int offset, const char *msg) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (skb->protocol == htons(ETH_P_IPV6)) + ah_esp_debug_packet_v6(pp, skb, offset, msg); + else +#endif + ah_esp_debug_packet_v4(pp, skb, offset, msg); +} + + +static void ah_esp_init(struct ip_vs_protocol *pp) +{ + /* nothing to do now */ +} + + +static void ah_esp_exit(struct ip_vs_protocol *pp) +{ + /* nothing to do now */ +} + + +#ifdef CONFIG_IP_VS_PROTO_AH +struct ip_vs_protocol ip_vs_protocol_ah = { + .name = "AH", + .protocol = IPPROTO_AH, + .num_states = 1, + .dont_defrag = 1, + .init = ah_esp_init, + .exit = ah_esp_exit, + .conn_schedule = ah_esp_conn_schedule, + .conn_in_get = ah_esp_conn_in_get, + .conn_out_get = ah_esp_conn_out_get, + .snat_handler = NULL, + .dnat_handler = NULL, + .csum_check = NULL, + .state_transition = NULL, + .register_app = NULL, + .unregister_app = NULL, + .app_conn_bind = NULL, + .debug_packet = ah_esp_debug_packet, + .timeout_change = NULL, /* ISAKMP */ + .set_state_timeout = NULL, +}; +#endif + +#ifdef CONFIG_IP_VS_PROTO_ESP +struct ip_vs_protocol ip_vs_protocol_esp = { + .name = "ESP", + .protocol = IPPROTO_ESP, + .num_states = 1, + .dont_defrag = 1, + .init = ah_esp_init, + .exit = ah_esp_exit, + .conn_schedule = ah_esp_conn_schedule, + .conn_in_get = ah_esp_conn_in_get, + .conn_out_get = ah_esp_conn_out_get, + .snat_handler = NULL, + .dnat_handler = NULL, + .csum_check = NULL, + .state_transition = NULL, + .register_app = NULL, + .unregister_app = NULL, + .app_conn_bind = NULL, + .debug_packet = ah_esp_debug_packet, + .timeout_change = NULL, /* ISAKMP */ +}; +#endif diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c new file mode 100644 index 00000000000..dd4566ea2bf --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -0,0 +1,732 @@ +/* + * ip_vs_proto_tcp.c: TCP load balancing support for IPVS + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +#include <linux/kernel.h> +#include <linux/ip.h> +#include <linux/tcp.h> /* for tcphdr */ +#include <net/ip.h> +#include <net/tcp.h> /* for csum_tcpudp_magic */ +#include <net/ip6_checksum.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> + +#include <net/ip_vs.h> + + +static struct ip_vs_conn * +tcp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, unsigned int proto_off, + int inverse) +{ + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + if (likely(!inverse)) { + return ip_vs_conn_in_get(af, iph->protocol, + &iph->saddr, pptr[0], + &iph->daddr, pptr[1]); + } else { + return ip_vs_conn_in_get(af, iph->protocol, + &iph->daddr, pptr[1], + &iph->saddr, pptr[0]); + } +} + +static struct ip_vs_conn * +tcp_conn_out_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, unsigned int proto_off, + int inverse) +{ + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + if (likely(!inverse)) { + return ip_vs_conn_out_get(af, iph->protocol, + &iph->saddr, pptr[0], + &iph->daddr, pptr[1]); + } else { + return ip_vs_conn_out_get(af, iph->protocol, + &iph->daddr, pptr[1], + &iph->saddr, pptr[0]); + } +} + + +static int +tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, + int *verdict, struct ip_vs_conn **cpp) +{ + struct ip_vs_service *svc; + struct tcphdr _tcph, *th; + struct ip_vs_iphdr iph; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + + th = skb_header_pointer(skb, iph.len, sizeof(_tcph), &_tcph); + if (th == NULL) { + *verdict = NF_DROP; + return 0; + } + + if (th->syn && + (svc = ip_vs_service_get(af, skb->mark, iph.protocol, &iph.daddr, + th->dest))) { + if (ip_vs_todrop()) { + /* + * It seems that we are very loaded. + * We have to drop this packet :( + */ + ip_vs_service_put(svc); + *verdict = NF_DROP; + return 0; + } + + /* + * Let the virtual server select a real server for the + * incoming connection, and create a connection entry. + */ + *cpp = ip_vs_schedule(svc, skb); + if (!*cpp) { + *verdict = ip_vs_leave(svc, skb, pp); + return 0; + } + ip_vs_service_put(svc); + } + return 1; +} + + +static inline void +tcp_fast_csum_update(int af, struct tcphdr *tcph, + const union nf_inet_addr *oldip, + const union nf_inet_addr *newip, + __be16 oldport, __be16 newport) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + tcph->check = + csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, + ip_vs_check_diff2(oldport, newport, + ~csum_unfold(tcph->check)))); + else +#endif + tcph->check = + csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, + ip_vs_check_diff2(oldport, newport, + ~csum_unfold(tcph->check)))); +} + + +static inline void +tcp_partial_csum_update(int af, struct tcphdr *tcph, + const union nf_inet_addr *oldip, + const union nf_inet_addr *newip, + __be16 oldlen, __be16 newlen) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + tcph->check = + csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, + ip_vs_check_diff2(oldlen, newlen, + ~csum_unfold(tcph->check)))); + else +#endif + tcph->check = + csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, + ip_vs_check_diff2(oldlen, newlen, + ~csum_unfold(tcph->check)))); +} + + +static int +tcp_snat_handler(struct sk_buff *skb, + struct ip_vs_protocol *pp, struct ip_vs_conn *cp) +{ + struct tcphdr *tcph; + unsigned int tcphoff; + int oldlen; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + tcphoff = sizeof(struct ipv6hdr); + else +#endif + tcphoff = ip_hdrlen(skb); + oldlen = skb->len - tcphoff; + + /* csum_check requires unshared skb */ + if (!skb_make_writable(skb, tcphoff+sizeof(*tcph))) + return 0; + + if (unlikely(cp->app != NULL)) { + /* Some checks before mangling */ + if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) + return 0; + + /* Call application helper if needed */ + if (!ip_vs_app_pkt_out(cp, skb)) + return 0; + } + + tcph = (void *)skb_network_header(skb) + tcphoff; + tcph->source = cp->vport; + + /* Adjust TCP checksums */ + if (skb->ip_summed == CHECKSUM_PARTIAL) { + tcp_partial_csum_update(cp->af, tcph, &cp->daddr, &cp->vaddr, + htonl(oldlen), + htonl(skb->len - tcphoff)); + } else if (!cp->app) { + /* Only port and addr are changed, do fast csum update */ + tcp_fast_csum_update(cp->af, tcph, &cp->daddr, &cp->vaddr, + cp->dport, cp->vport); + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->ip_summed = CHECKSUM_NONE; + } else { + /* full checksum calculation */ + tcph->check = 0; + skb->csum = skb_checksum(skb, tcphoff, skb->len - tcphoff, 0); +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + tcph->check = csum_ipv6_magic(&cp->vaddr.in6, + &cp->caddr.in6, + skb->len - tcphoff, + cp->protocol, skb->csum); + else +#endif + tcph->check = csum_tcpudp_magic(cp->vaddr.ip, + cp->caddr.ip, + skb->len - tcphoff, + cp->protocol, + skb->csum); + + IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", + pp->name, tcph->check, + (char*)&(tcph->check) - (char*)tcph); + } + return 1; +} + + +static int +tcp_dnat_handler(struct sk_buff *skb, + struct ip_vs_protocol *pp, struct ip_vs_conn *cp) +{ + struct tcphdr *tcph; + unsigned int tcphoff; + int oldlen; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + tcphoff = sizeof(struct ipv6hdr); + else +#endif + tcphoff = ip_hdrlen(skb); + oldlen = skb->len - tcphoff; + + /* csum_check requires unshared skb */ + if (!skb_make_writable(skb, tcphoff+sizeof(*tcph))) + return 0; + + if (unlikely(cp->app != NULL)) { + /* Some checks before mangling */ + if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) + return 0; + + /* + * Attempt ip_vs_app call. + * It will fix ip_vs_conn and iph ack_seq stuff + */ + if (!ip_vs_app_pkt_in(cp, skb)) + return 0; + } + + tcph = (void *)skb_network_header(skb) + tcphoff; + tcph->dest = cp->dport; + + /* + * Adjust TCP checksums + */ + if (skb->ip_summed == CHECKSUM_PARTIAL) { + tcp_partial_csum_update(cp->af, tcph, &cp->daddr, &cp->vaddr, + htonl(oldlen), + htonl(skb->len - tcphoff)); + } else if (!cp->app) { + /* Only port and addr are changed, do fast csum update */ + tcp_fast_csum_update(cp->af, tcph, &cp->vaddr, &cp->daddr, + cp->vport, cp->dport); + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->ip_summed = CHECKSUM_NONE; + } else { + /* full checksum calculation */ + tcph->check = 0; + skb->csum = skb_checksum(skb, tcphoff, skb->len - tcphoff, 0); +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + tcph->check = csum_ipv6_magic(&cp->caddr.in6, + &cp->daddr.in6, + skb->len - tcphoff, + cp->protocol, skb->csum); + else +#endif + tcph->check = csum_tcpudp_magic(cp->caddr.ip, + cp->daddr.ip, + skb->len - tcphoff, + cp->protocol, + skb->csum); + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + return 1; +} + + +static int +tcp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) +{ + unsigned int tcphoff; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + tcphoff = sizeof(struct ipv6hdr); + else +#endif + tcphoff = ip_hdrlen(skb); + + switch (skb->ip_summed) { + case CHECKSUM_NONE: + skb->csum = skb_checksum(skb, tcphoff, skb->len - tcphoff, 0); + case CHECKSUM_COMPLETE: +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + skb->len - tcphoff, + ipv6_hdr(skb)->nexthdr, + skb->csum)) { + IP_VS_DBG_RL_PKT(0, pp, skb, 0, + "Failed checksum for"); + return 0; + } + } else +#endif + if (csum_tcpudp_magic(ip_hdr(skb)->saddr, + ip_hdr(skb)->daddr, + skb->len - tcphoff, + ip_hdr(skb)->protocol, + skb->csum)) { + IP_VS_DBG_RL_PKT(0, pp, skb, 0, + "Failed checksum for"); + return 0; + } + break; + default: + /* No need to checksum. */ + break; + } + + return 1; +} + + +#define TCP_DIR_INPUT 0 +#define TCP_DIR_OUTPUT 4 +#define TCP_DIR_INPUT_ONLY 8 + +static const int tcp_state_off[IP_VS_DIR_LAST] = { + [IP_VS_DIR_INPUT] = TCP_DIR_INPUT, + [IP_VS_DIR_OUTPUT] = TCP_DIR_OUTPUT, + [IP_VS_DIR_INPUT_ONLY] = TCP_DIR_INPUT_ONLY, +}; + +/* + * Timeout table[state] + */ +static int tcp_timeouts[IP_VS_TCP_S_LAST+1] = { + [IP_VS_TCP_S_NONE] = 2*HZ, + [IP_VS_TCP_S_ESTABLISHED] = 15*60*HZ, + [IP_VS_TCP_S_SYN_SENT] = 2*60*HZ, + [IP_VS_TCP_S_SYN_RECV] = 1*60*HZ, + [IP_VS_TCP_S_FIN_WAIT] = 2*60*HZ, + [IP_VS_TCP_S_TIME_WAIT] = 2*60*HZ, + [IP_VS_TCP_S_CLOSE] = 10*HZ, + [IP_VS_TCP_S_CLOSE_WAIT] = 60*HZ, + [IP_VS_TCP_S_LAST_ACK] = 30*HZ, + [IP_VS_TCP_S_LISTEN] = 2*60*HZ, + [IP_VS_TCP_S_SYNACK] = 120*HZ, + [IP_VS_TCP_S_LAST] = 2*HZ, +}; + +static char * tcp_state_name_table[IP_VS_TCP_S_LAST+1] = { + [IP_VS_TCP_S_NONE] = "NONE", + [IP_VS_TCP_S_ESTABLISHED] = "ESTABLISHED", + [IP_VS_TCP_S_SYN_SENT] = "SYN_SENT", + [IP_VS_TCP_S_SYN_RECV] = "SYN_RECV", + [IP_VS_TCP_S_FIN_WAIT] = "FIN_WAIT", + [IP_VS_TCP_S_TIME_WAIT] = "TIME_WAIT", + [IP_VS_TCP_S_CLOSE] = "CLOSE", + [IP_VS_TCP_S_CLOSE_WAIT] = "CLOSE_WAIT", + [IP_VS_TCP_S_LAST_ACK] = "LAST_ACK", + [IP_VS_TCP_S_LISTEN] = "LISTEN", + [IP_VS_TCP_S_SYNACK] = "SYNACK", + [IP_VS_TCP_S_LAST] = "BUG!", +}; + +#define sNO IP_VS_TCP_S_NONE +#define sES IP_VS_TCP_S_ESTABLISHED +#define sSS IP_VS_TCP_S_SYN_SENT +#define sSR IP_VS_TCP_S_SYN_RECV +#define sFW IP_VS_TCP_S_FIN_WAIT +#define sTW IP_VS_TCP_S_TIME_WAIT +#define sCL IP_VS_TCP_S_CLOSE +#define sCW IP_VS_TCP_S_CLOSE_WAIT +#define sLA IP_VS_TCP_S_LAST_ACK +#define sLI IP_VS_TCP_S_LISTEN +#define sSA IP_VS_TCP_S_SYNACK + +struct tcp_states_t { + int next_state[IP_VS_TCP_S_LAST]; +}; + +static const char * tcp_state_name(int state) +{ + if (state >= IP_VS_TCP_S_LAST) + return "ERR!"; + return tcp_state_name_table[state] ? tcp_state_name_table[state] : "?"; +} + +static struct tcp_states_t tcp_states [] = { +/* INPUT */ +/* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ +/*syn*/ {{sSR, sES, sES, sSR, sSR, sSR, sSR, sSR, sSR, sSR, sSR }}, +/*fin*/ {{sCL, sCW, sSS, sTW, sTW, sTW, sCL, sCW, sLA, sLI, sTW }}, +/*ack*/ {{sCL, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }}, +/*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sSR }}, + +/* OUTPUT */ +/* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ +/*syn*/ {{sSS, sES, sSS, sSR, sSS, sSS, sSS, sSS, sSS, sLI, sSR }}, +/*fin*/ {{sTW, sFW, sSS, sTW, sFW, sTW, sCL, sTW, sLA, sLI, sTW }}, +/*ack*/ {{sES, sES, sSS, sES, sFW, sTW, sCL, sCW, sLA, sES, sES }}, +/*rst*/ {{sCL, sCL, sSS, sCL, sCL, sTW, sCL, sCL, sCL, sCL, sCL }}, + +/* INPUT-ONLY */ +/* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ +/*syn*/ {{sSR, sES, sES, sSR, sSR, sSR, sSR, sSR, sSR, sSR, sSR }}, +/*fin*/ {{sCL, sFW, sSS, sTW, sFW, sTW, sCL, sCW, sLA, sLI, sTW }}, +/*ack*/ {{sCL, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }}, +/*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }}, +}; + +static struct tcp_states_t tcp_states_dos [] = { +/* INPUT */ +/* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ +/*syn*/ {{sSR, sES, sES, sSR, sSR, sSR, sSR, sSR, sSR, sSR, sSA }}, +/*fin*/ {{sCL, sCW, sSS, sTW, sTW, sTW, sCL, sCW, sLA, sLI, sSA }}, +/*ack*/ {{sCL, sES, sSS, sSR, sFW, sTW, sCL, sCW, sCL, sLI, sSA }}, +/*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }}, + +/* OUTPUT */ +/* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ +/*syn*/ {{sSS, sES, sSS, sSA, sSS, sSS, sSS, sSS, sSS, sLI, sSA }}, +/*fin*/ {{sTW, sFW, sSS, sTW, sFW, sTW, sCL, sTW, sLA, sLI, sTW }}, +/*ack*/ {{sES, sES, sSS, sES, sFW, sTW, sCL, sCW, sLA, sES, sES }}, +/*rst*/ {{sCL, sCL, sSS, sCL, sCL, sTW, sCL, sCL, sCL, sCL, sCL }}, + +/* INPUT-ONLY */ +/* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ +/*syn*/ {{sSA, sES, sES, sSR, sSA, sSA, sSA, sSA, sSA, sSA, sSA }}, +/*fin*/ {{sCL, sFW, sSS, sTW, sFW, sTW, sCL, sCW, sLA, sLI, sTW }}, +/*ack*/ {{sCL, sES, sSS, sES, sFW, sTW, sCL, sCW, sCL, sLI, sES }}, +/*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }}, +}; + +static struct tcp_states_t *tcp_state_table = tcp_states; + + +static void tcp_timeout_change(struct ip_vs_protocol *pp, int flags) +{ + int on = (flags & 1); /* secure_tcp */ + + /* + ** FIXME: change secure_tcp to independent sysctl var + ** or make it per-service or per-app because it is valid + ** for most if not for all of the applications. Something + ** like "capabilities" (flags) for each object. + */ + tcp_state_table = (on? tcp_states_dos : tcp_states); +} + +static int +tcp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) +{ + return ip_vs_set_state_timeout(pp->timeout_table, IP_VS_TCP_S_LAST, + tcp_state_name_table, sname, to); +} + +static inline int tcp_state_idx(struct tcphdr *th) +{ + if (th->rst) + return 3; + if (th->syn) + return 0; + if (th->fin) + return 1; + if (th->ack) + return 2; + return -1; +} + +static inline void +set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, + int direction, struct tcphdr *th) +{ + int state_idx; + int new_state = IP_VS_TCP_S_CLOSE; + int state_off = tcp_state_off[direction]; + + /* + * Update state offset to INPUT_ONLY if necessary + * or delete NO_OUTPUT flag if output packet detected + */ + if (cp->flags & IP_VS_CONN_F_NOOUTPUT) { + if (state_off == TCP_DIR_OUTPUT) + cp->flags &= ~IP_VS_CONN_F_NOOUTPUT; + else + state_off = TCP_DIR_INPUT_ONLY; + } + + if ((state_idx = tcp_state_idx(th)) < 0) { + IP_VS_DBG(8, "tcp_state_idx=%d!!!\n", state_idx); + goto tcp_state_out; + } + + new_state = tcp_state_table[state_off+state_idx].next_state[cp->state]; + + tcp_state_out: + if (new_state != cp->state) { + struct ip_vs_dest *dest = cp->dest; + + IP_VS_DBG_BUF(8, "%s %s [%c%c%c%c] %s:%d->" + "%s:%d state: %s->%s conn->refcnt:%d\n", + pp->name, + ((state_off == TCP_DIR_OUTPUT) ? + "output " : "input "), + th->syn ? 'S' : '.', + th->fin ? 'F' : '.', + th->ack ? 'A' : '.', + th->rst ? 'R' : '.', + IP_VS_DBG_ADDR(cp->af, &cp->daddr), + ntohs(cp->dport), + IP_VS_DBG_ADDR(cp->af, &cp->caddr), + ntohs(cp->cport), + tcp_state_name(cp->state), + tcp_state_name(new_state), + atomic_read(&cp->refcnt)); + + if (dest) { + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (new_state != IP_VS_TCP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags |= IP_VS_CONN_F_INACTIVE; + } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && + (new_state == IP_VS_TCP_S_ESTABLISHED)) { + atomic_inc(&dest->activeconns); + atomic_dec(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } + } + } + + cp->timeout = pp->timeout_table[cp->state = new_state]; +} + + +/* + * Handle state transitions + */ +static int +tcp_state_transition(struct ip_vs_conn *cp, int direction, + const struct sk_buff *skb, + struct ip_vs_protocol *pp) +{ + struct tcphdr _tcph, *th; + +#ifdef CONFIG_IP_VS_IPV6 + int ihl = cp->af == AF_INET ? ip_hdrlen(skb) : sizeof(struct ipv6hdr); +#else + int ihl = ip_hdrlen(skb); +#endif + + th = skb_header_pointer(skb, ihl, sizeof(_tcph), &_tcph); + if (th == NULL) + return 0; + + spin_lock(&cp->lock); + set_tcp_state(pp, cp, direction, th); + spin_unlock(&cp->lock); + + return 1; +} + + +/* + * Hash table for TCP application incarnations + */ +#define TCP_APP_TAB_BITS 4 +#define TCP_APP_TAB_SIZE (1 << TCP_APP_TAB_BITS) +#define TCP_APP_TAB_MASK (TCP_APP_TAB_SIZE - 1) + +static struct list_head tcp_apps[TCP_APP_TAB_SIZE]; +static DEFINE_SPINLOCK(tcp_app_lock); + +static inline __u16 tcp_app_hashkey(__be16 port) +{ + return (((__force u16)port >> TCP_APP_TAB_BITS) ^ (__force u16)port) + & TCP_APP_TAB_MASK; +} + + +static int tcp_register_app(struct ip_vs_app *inc) +{ + struct ip_vs_app *i; + __u16 hash; + __be16 port = inc->port; + int ret = 0; + + hash = tcp_app_hashkey(port); + + spin_lock_bh(&tcp_app_lock); + list_for_each_entry(i, &tcp_apps[hash], p_list) { + if (i->port == port) { + ret = -EEXIST; + goto out; + } + } + list_add(&inc->p_list, &tcp_apps[hash]); + atomic_inc(&ip_vs_protocol_tcp.appcnt); + + out: + spin_unlock_bh(&tcp_app_lock); + return ret; +} + + +static void +tcp_unregister_app(struct ip_vs_app *inc) +{ + spin_lock_bh(&tcp_app_lock); + atomic_dec(&ip_vs_protocol_tcp.appcnt); + list_del(&inc->p_list); + spin_unlock_bh(&tcp_app_lock); +} + + +static int +tcp_app_conn_bind(struct ip_vs_conn *cp) +{ + int hash; + struct ip_vs_app *inc; + int result = 0; + + /* Default binding: bind app only for NAT */ + if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) + return 0; + + /* Lookup application incarnations and bind the right one */ + hash = tcp_app_hashkey(cp->vport); + + spin_lock(&tcp_app_lock); + list_for_each_entry(inc, &tcp_apps[hash], p_list) { + if (inc->port == cp->vport) { + if (unlikely(!ip_vs_app_inc_get(inc))) + break; + spin_unlock(&tcp_app_lock); + + IP_VS_DBG_BUF(9, "%s: Binding conn %s:%u->" + "%s:%u to app %s on port %u\n", + __func__, + IP_VS_DBG_ADDR(cp->af, &cp->caddr), + ntohs(cp->cport), + IP_VS_DBG_ADDR(cp->af, &cp->vaddr), + ntohs(cp->vport), + inc->name, ntohs(inc->port)); + + cp->app = inc; + if (inc->init_conn) + result = inc->init_conn(inc, cp); + goto out; + } + } + spin_unlock(&tcp_app_lock); + + out: + return result; +} + + +/* + * Set LISTEN timeout. (ip_vs_conn_put will setup timer) + */ +void ip_vs_tcp_conn_listen(struct ip_vs_conn *cp) +{ + spin_lock(&cp->lock); + cp->state = IP_VS_TCP_S_LISTEN; + cp->timeout = ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_LISTEN]; + spin_unlock(&cp->lock); +} + + +static void ip_vs_tcp_init(struct ip_vs_protocol *pp) +{ + IP_VS_INIT_HASH_TABLE(tcp_apps); + pp->timeout_table = tcp_timeouts; +} + + +static void ip_vs_tcp_exit(struct ip_vs_protocol *pp) +{ +} + + +struct ip_vs_protocol ip_vs_protocol_tcp = { + .name = "TCP", + .protocol = IPPROTO_TCP, + .num_states = IP_VS_TCP_S_LAST, + .dont_defrag = 0, + .appcnt = ATOMIC_INIT(0), + .init = ip_vs_tcp_init, + .exit = ip_vs_tcp_exit, + .register_app = tcp_register_app, + .unregister_app = tcp_unregister_app, + .conn_schedule = tcp_conn_schedule, + .conn_in_get = tcp_conn_in_get, + .conn_out_get = tcp_conn_out_get, + .snat_handler = tcp_snat_handler, + .dnat_handler = tcp_dnat_handler, + .csum_check = tcp_csum_check, + .state_name = tcp_state_name, + .state_transition = tcp_state_transition, + .app_conn_bind = tcp_app_conn_bind, + .debug_packet = ip_vs_tcpudp_debug_packet, + .timeout_change = tcp_timeout_change, + .set_state_timeout = tcp_set_state_timeout, +}; diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c new file mode 100644 index 00000000000..6eb6039d634 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -0,0 +1,533 @@ +/* + * ip_vs_proto_udp.c: UDP load balancing support for IPVS + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +#include <linux/in.h> +#include <linux/ip.h> +#include <linux/kernel.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> +#include <linux/udp.h> + +#include <net/ip_vs.h> +#include <net/ip.h> +#include <net/ip6_checksum.h> + +static struct ip_vs_conn * +udp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, unsigned int proto_off, + int inverse) +{ + struct ip_vs_conn *cp; + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + if (likely(!inverse)) { + cp = ip_vs_conn_in_get(af, iph->protocol, + &iph->saddr, pptr[0], + &iph->daddr, pptr[1]); + } else { + cp = ip_vs_conn_in_get(af, iph->protocol, + &iph->daddr, pptr[1], + &iph->saddr, pptr[0]); + } + + return cp; +} + + +static struct ip_vs_conn * +udp_conn_out_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, + const struct ip_vs_iphdr *iph, unsigned int proto_off, + int inverse) +{ + struct ip_vs_conn *cp; + __be16 _ports[2], *pptr; + + pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); + if (pptr == NULL) + return NULL; + + if (likely(!inverse)) { + cp = ip_vs_conn_out_get(af, iph->protocol, + &iph->saddr, pptr[0], + &iph->daddr, pptr[1]); + } else { + cp = ip_vs_conn_out_get(af, iph->protocol, + &iph->daddr, pptr[1], + &iph->saddr, pptr[0]); + } + + return cp; +} + + +static int +udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, + int *verdict, struct ip_vs_conn **cpp) +{ + struct ip_vs_service *svc; + struct udphdr _udph, *uh; + struct ip_vs_iphdr iph; + + ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); + + uh = skb_header_pointer(skb, iph.len, sizeof(_udph), &_udph); + if (uh == NULL) { + *verdict = NF_DROP; + return 0; + } + + svc = ip_vs_service_get(af, skb->mark, iph.protocol, + &iph.daddr, uh->dest); + if (svc) { + if (ip_vs_todrop()) { + /* + * It seems that we are very loaded. + * We have to drop this packet :( + */ + ip_vs_service_put(svc); + *verdict = NF_DROP; + return 0; + } + + /* + * Let the virtual server select a real server for the + * incoming connection, and create a connection entry. + */ + *cpp = ip_vs_schedule(svc, skb); + if (!*cpp) { + *verdict = ip_vs_leave(svc, skb, pp); + return 0; + } + ip_vs_service_put(svc); + } + return 1; +} + + +static inline void +udp_fast_csum_update(int af, struct udphdr *uhdr, + const union nf_inet_addr *oldip, + const union nf_inet_addr *newip, + __be16 oldport, __be16 newport) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + uhdr->check = + csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, + ip_vs_check_diff2(oldport, newport, + ~csum_unfold(uhdr->check)))); + else +#endif + uhdr->check = + csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, + ip_vs_check_diff2(oldport, newport, + ~csum_unfold(uhdr->check)))); + if (!uhdr->check) + uhdr->check = CSUM_MANGLED_0; +} + +static inline void +udp_partial_csum_update(int af, struct udphdr *uhdr, + const union nf_inet_addr *oldip, + const union nf_inet_addr *newip, + __be16 oldlen, __be16 newlen) +{ +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + uhdr->check = + csum_fold(ip_vs_check_diff16(oldip->ip6, newip->ip6, + ip_vs_check_diff2(oldlen, newlen, + ~csum_unfold(uhdr->check)))); + else +#endif + uhdr->check = + csum_fold(ip_vs_check_diff4(oldip->ip, newip->ip, + ip_vs_check_diff2(oldlen, newlen, + ~csum_unfold(uhdr->check)))); +} + + +static int +udp_snat_handler(struct sk_buff *skb, + struct ip_vs_protocol *pp, struct ip_vs_conn *cp) +{ + struct udphdr *udph; + unsigned int udphoff; + int oldlen; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + udphoff = sizeof(struct ipv6hdr); + else +#endif + udphoff = ip_hdrlen(skb); + oldlen = skb->len - udphoff; + + /* csum_check requires unshared skb */ + if (!skb_make_writable(skb, udphoff+sizeof(*udph))) + return 0; + + if (unlikely(cp->app != NULL)) { + /* Some checks before mangling */ + if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) + return 0; + + /* + * Call application helper if needed + */ + if (!ip_vs_app_pkt_out(cp, skb)) + return 0; + } + + udph = (void *)skb_network_header(skb) + udphoff; + udph->source = cp->vport; + + /* + * Adjust UDP checksums + */ + if (skb->ip_summed == CHECKSUM_PARTIAL) { + udp_partial_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, + htonl(oldlen), + htonl(skb->len - udphoff)); + } else if (!cp->app && (udph->check != 0)) { + /* Only port and addr are changed, do fast csum update */ + udp_fast_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, + cp->dport, cp->vport); + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->ip_summed = CHECKSUM_NONE; + } else { + /* full checksum calculation */ + udph->check = 0; + skb->csum = skb_checksum(skb, udphoff, skb->len - udphoff, 0); +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + udph->check = csum_ipv6_magic(&cp->vaddr.in6, + &cp->caddr.in6, + skb->len - udphoff, + cp->protocol, skb->csum); + else +#endif + udph->check = csum_tcpudp_magic(cp->vaddr.ip, + cp->caddr.ip, + skb->len - udphoff, + cp->protocol, + skb->csum); + if (udph->check == 0) + udph->check = CSUM_MANGLED_0; + IP_VS_DBG(11, "O-pkt: %s O-csum=%d (+%zd)\n", + pp->name, udph->check, + (char*)&(udph->check) - (char*)udph); + } + return 1; +} + + +static int +udp_dnat_handler(struct sk_buff *skb, + struct ip_vs_protocol *pp, struct ip_vs_conn *cp) +{ + struct udphdr *udph; + unsigned int udphoff; + int oldlen; + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + udphoff = sizeof(struct ipv6hdr); + else +#endif + udphoff = ip_hdrlen(skb); + oldlen = skb->len - udphoff; + + /* csum_check requires unshared skb */ + if (!skb_make_writable(skb, udphoff+sizeof(*udph))) + return 0; + + if (unlikely(cp->app != NULL)) { + /* Some checks before mangling */ + if (pp->csum_check && !pp->csum_check(cp->af, skb, pp)) + return 0; + + /* + * Attempt ip_vs_app call. + * It will fix ip_vs_conn + */ + if (!ip_vs_app_pkt_in(cp, skb)) + return 0; + } + + udph = (void *)skb_network_header(skb) + udphoff; + udph->dest = cp->dport; + + /* + * Adjust UDP checksums + */ + if (skb->ip_summed == CHECKSUM_PARTIAL) { + udp_partial_csum_update(cp->af, udph, &cp->daddr, &cp->vaddr, + htonl(oldlen), + htonl(skb->len - udphoff)); + } else if (!cp->app && (udph->check != 0)) { + /* Only port and addr are changed, do fast csum update */ + udp_fast_csum_update(cp->af, udph, &cp->vaddr, &cp->daddr, + cp->vport, cp->dport); + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->ip_summed = CHECKSUM_NONE; + } else { + /* full checksum calculation */ + udph->check = 0; + skb->csum = skb_checksum(skb, udphoff, skb->len - udphoff, 0); +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + udph->check = csum_ipv6_magic(&cp->caddr.in6, + &cp->daddr.in6, + skb->len - udphoff, + cp->protocol, skb->csum); + else +#endif + udph->check = csum_tcpudp_magic(cp->caddr.ip, + cp->daddr.ip, + skb->len - udphoff, + cp->protocol, + skb->csum); + if (udph->check == 0) + udph->check = CSUM_MANGLED_0; + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + return 1; +} + + +static int +udp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) +{ + struct udphdr _udph, *uh; + unsigned int udphoff; + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + udphoff = sizeof(struct ipv6hdr); + else +#endif + udphoff = ip_hdrlen(skb); + + uh = skb_header_pointer(skb, udphoff, sizeof(_udph), &_udph); + if (uh == NULL) + return 0; + + if (uh->check != 0) { + switch (skb->ip_summed) { + case CHECKSUM_NONE: + skb->csum = skb_checksum(skb, udphoff, + skb->len - udphoff, 0); + case CHECKSUM_COMPLETE: +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + skb->len - udphoff, + ipv6_hdr(skb)->nexthdr, + skb->csum)) { + IP_VS_DBG_RL_PKT(0, pp, skb, 0, + "Failed checksum for"); + return 0; + } + } else +#endif + if (csum_tcpudp_magic(ip_hdr(skb)->saddr, + ip_hdr(skb)->daddr, + skb->len - udphoff, + ip_hdr(skb)->protocol, + skb->csum)) { + IP_VS_DBG_RL_PKT(0, pp, skb, 0, + "Failed checksum for"); + return 0; + } + break; + default: + /* No need to checksum. */ + break; + } + } + return 1; +} + + +/* + * Note: the caller guarantees that only one of register_app, + * unregister_app or app_conn_bind is called each time. + */ + +#define UDP_APP_TAB_BITS 4 +#define UDP_APP_TAB_SIZE (1 << UDP_APP_TAB_BITS) +#define UDP_APP_TAB_MASK (UDP_APP_TAB_SIZE - 1) + +static struct list_head udp_apps[UDP_APP_TAB_SIZE]; +static DEFINE_SPINLOCK(udp_app_lock); + +static inline __u16 udp_app_hashkey(__be16 port) +{ + return (((__force u16)port >> UDP_APP_TAB_BITS) ^ (__force u16)port) + & UDP_APP_TAB_MASK; +} + + +static int udp_register_app(struct ip_vs_app *inc) +{ + struct ip_vs_app *i; + __u16 hash; + __be16 port = inc->port; + int ret = 0; + + hash = udp_app_hashkey(port); + + + spin_lock_bh(&udp_app_lock); + list_for_each_entry(i, &udp_apps[hash], p_list) { + if (i->port == port) { + ret = -EEXIST; + goto out; + } + } + list_add(&inc->p_list, &udp_apps[hash]); + atomic_inc(&ip_vs_protocol_udp.appcnt); + + out: + spin_unlock_bh(&udp_app_lock); + return ret; +} + + +static void +udp_unregister_app(struct ip_vs_app *inc) +{ + spin_lock_bh(&udp_app_lock); + atomic_dec(&ip_vs_protocol_udp.appcnt); + list_del(&inc->p_list); + spin_unlock_bh(&udp_app_lock); +} + + +static int udp_app_conn_bind(struct ip_vs_conn *cp) +{ + int hash; + struct ip_vs_app *inc; + int result = 0; + + /* Default binding: bind app only for NAT */ + if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) + return 0; + + /* Lookup application incarnations and bind the right one */ + hash = udp_app_hashkey(cp->vport); + + spin_lock(&udp_app_lock); + list_for_each_entry(inc, &udp_apps[hash], p_list) { + if (inc->port == cp->vport) { + if (unlikely(!ip_vs_app_inc_get(inc))) + break; + spin_unlock(&udp_app_lock); + + IP_VS_DBG_BUF(9, "%s: Binding conn %s:%u->" + "%s:%u to app %s on port %u\n", + __func__, + IP_VS_DBG_ADDR(cp->af, &cp->caddr), + ntohs(cp->cport), + IP_VS_DBG_ADDR(cp->af, &cp->vaddr), + ntohs(cp->vport), + inc->name, ntohs(inc->port)); + + cp->app = inc; + if (inc->init_conn) + result = inc->init_conn(inc, cp); + goto out; + } + } + spin_unlock(&udp_app_lock); + + out: + return result; +} + + +static int udp_timeouts[IP_VS_UDP_S_LAST+1] = { + [IP_VS_UDP_S_NORMAL] = 5*60*HZ, + [IP_VS_UDP_S_LAST] = 2*HZ, +}; + +static char * udp_state_name_table[IP_VS_UDP_S_LAST+1] = { + [IP_VS_UDP_S_NORMAL] = "UDP", + [IP_VS_UDP_S_LAST] = "BUG!", +}; + + +static int +udp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) +{ + return ip_vs_set_state_timeout(pp->timeout_table, IP_VS_UDP_S_LAST, + udp_state_name_table, sname, to); +} + +static const char * udp_state_name(int state) +{ + if (state >= IP_VS_UDP_S_LAST) + return "ERR!"; + return udp_state_name_table[state] ? udp_state_name_table[state] : "?"; +} + +static int +udp_state_transition(struct ip_vs_conn *cp, int direction, + const struct sk_buff *skb, + struct ip_vs_protocol *pp) +{ + cp->timeout = pp->timeout_table[IP_VS_UDP_S_NORMAL]; + return 1; +} + +static void udp_init(struct ip_vs_protocol *pp) +{ + IP_VS_INIT_HASH_TABLE(udp_apps); + pp->timeout_table = udp_timeouts; +} + +static void udp_exit(struct ip_vs_protocol *pp) +{ +} + + +struct ip_vs_protocol ip_vs_protocol_udp = { + .name = "UDP", + .protocol = IPPROTO_UDP, + .num_states = IP_VS_UDP_S_LAST, + .dont_defrag = 0, + .init = udp_init, + .exit = udp_exit, + .conn_schedule = udp_conn_schedule, + .conn_in_get = udp_conn_in_get, + .conn_out_get = udp_conn_out_get, + .snat_handler = udp_snat_handler, + .dnat_handler = udp_dnat_handler, + .csum_check = udp_csum_check, + .state_transition = udp_state_transition, + .state_name = udp_state_name, + .register_app = udp_register_app, + .unregister_app = udp_unregister_app, + .app_conn_bind = udp_app_conn_bind, + .debug_packet = ip_vs_tcpudp_debug_packet, + .timeout_change = NULL, + .set_state_timeout = udp_set_state_timeout, +}; diff --git a/net/netfilter/ipvs/ip_vs_rr.c b/net/netfilter/ipvs/ip_vs_rr.c new file mode 100644 index 00000000000..a22195f68ac --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_rr.c @@ -0,0 +1,112 @@ +/* + * IPVS: Round-Robin Scheduling module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Peter Kese <peter.kese@ijs.si> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Fixes/Changes: + * Wensong Zhang : changed the ip_vs_rr_schedule to return dest + * Julian Anastasov : fixed the NULL pointer access bug in debugging + * Wensong Zhang : changed some comestics things for debugging + * Wensong Zhang : changed for the d-linked destination list + * Wensong Zhang : added the ip_vs_rr_update_svc + * Wensong Zhang : added any dest with weight=0 is quiesced + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> + +#include <net/ip_vs.h> + + +static int ip_vs_rr_init_svc(struct ip_vs_service *svc) +{ + svc->sched_data = &svc->destinations; + return 0; +} + + +static int ip_vs_rr_update_svc(struct ip_vs_service *svc) +{ + svc->sched_data = &svc->destinations; + return 0; +} + + +/* + * Round-Robin Scheduling + */ +static struct ip_vs_dest * +ip_vs_rr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct list_head *p, *q; + struct ip_vs_dest *dest; + + IP_VS_DBG(6, "ip_vs_rr_schedule(): Scheduling...\n"); + + write_lock(&svc->sched_lock); + p = (struct list_head *)svc->sched_data; + p = p->next; + q = p; + do { + /* skip list head */ + if (q == &svc->destinations) { + q = q->next; + continue; + } + + dest = list_entry(q, struct ip_vs_dest, n_list); + if (!(dest->flags & IP_VS_DEST_F_OVERLOAD) && + atomic_read(&dest->weight) > 0) + /* HIT */ + goto out; + q = q->next; + } while (q != p); + write_unlock(&svc->sched_lock); + return NULL; + + out: + svc->sched_data = q; + write_unlock(&svc->sched_lock); + IP_VS_DBG_BUF(6, "RR: server %s:%u " + "activeconns %d refcnt %d weight %d\n", + IP_VS_DBG_ADDR(svc->af, &dest->addr), ntohs(dest->port), + atomic_read(&dest->activeconns), + atomic_read(&dest->refcnt), atomic_read(&dest->weight)); + + return dest; +} + + +static struct ip_vs_scheduler ip_vs_rr_scheduler = { + .name = "rr", /* name */ + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_rr_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 1, +#endif + .init_service = ip_vs_rr_init_svc, + .update_service = ip_vs_rr_update_svc, + .schedule = ip_vs_rr_schedule, +}; + +static int __init ip_vs_rr_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_rr_scheduler); +} + +static void __exit ip_vs_rr_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_rr_scheduler); +} + +module_init(ip_vs_rr_init); +module_exit(ip_vs_rr_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_sched.c b/net/netfilter/ipvs/ip_vs_sched.c new file mode 100644 index 00000000000..a46ad9e3501 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_sched.c @@ -0,0 +1,251 @@ +/* + * IPVS An implementation of the IP virtual server support for the + * LINUX operating system. IPVS is now implemented as a module + * over the Netfilter framework. IPVS can be used to build a + * high-performance and highly available server based on a + * cluster of servers. + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Peter Kese <peter.kese@ijs.si> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +#include <linux/module.h> +#include <linux/spinlock.h> +#include <linux/interrupt.h> +#include <asm/string.h> +#include <linux/kmod.h> +#include <linux/sysctl.h> + +#include <net/ip_vs.h> + +/* + * IPVS scheduler list + */ +static LIST_HEAD(ip_vs_schedulers); + +/* lock for service table */ +static DEFINE_RWLOCK(__ip_vs_sched_lock); + + +/* + * Bind a service with a scheduler + */ +int ip_vs_bind_scheduler(struct ip_vs_service *svc, + struct ip_vs_scheduler *scheduler) +{ + int ret; + + if (svc == NULL) { + IP_VS_ERR("ip_vs_bind_scheduler(): svc arg NULL\n"); + return -EINVAL; + } + if (scheduler == NULL) { + IP_VS_ERR("ip_vs_bind_scheduler(): scheduler arg NULL\n"); + return -EINVAL; + } + + svc->scheduler = scheduler; + + if (scheduler->init_service) { + ret = scheduler->init_service(svc); + if (ret) { + IP_VS_ERR("ip_vs_bind_scheduler(): init error\n"); + return ret; + } + } + + return 0; +} + + +/* + * Unbind a service with its scheduler + */ +int ip_vs_unbind_scheduler(struct ip_vs_service *svc) +{ + struct ip_vs_scheduler *sched; + + if (svc == NULL) { + IP_VS_ERR("ip_vs_unbind_scheduler(): svc arg NULL\n"); + return -EINVAL; + } + + sched = svc->scheduler; + if (sched == NULL) { + IP_VS_ERR("ip_vs_unbind_scheduler(): svc isn't bound\n"); + return -EINVAL; + } + + if (sched->done_service) { + if (sched->done_service(svc) != 0) { + IP_VS_ERR("ip_vs_unbind_scheduler(): done error\n"); + return -EINVAL; + } + } + + svc->scheduler = NULL; + return 0; +} + + +/* + * Get scheduler in the scheduler list by name + */ +static struct ip_vs_scheduler *ip_vs_sched_getbyname(const char *sched_name) +{ + struct ip_vs_scheduler *sched; + + IP_VS_DBG(2, "ip_vs_sched_getbyname(): sched_name \"%s\"\n", + sched_name); + + read_lock_bh(&__ip_vs_sched_lock); + + list_for_each_entry(sched, &ip_vs_schedulers, n_list) { + /* + * Test and get the modules atomically + */ + if (sched->module && !try_module_get(sched->module)) { + /* + * This scheduler is just deleted + */ + continue; + } + if (strcmp(sched_name, sched->name)==0) { + /* HIT */ + read_unlock_bh(&__ip_vs_sched_lock); + return sched; + } + if (sched->module) + module_put(sched->module); + } + + read_unlock_bh(&__ip_vs_sched_lock); + return NULL; +} + + +/* + * Lookup scheduler and try to load it if it doesn't exist + */ +struct ip_vs_scheduler *ip_vs_scheduler_get(const char *sched_name) +{ + struct ip_vs_scheduler *sched; + + /* + * Search for the scheduler by sched_name + */ + sched = ip_vs_sched_getbyname(sched_name); + + /* + * If scheduler not found, load the module and search again + */ + if (sched == NULL) { + request_module("ip_vs_%s", sched_name); + sched = ip_vs_sched_getbyname(sched_name); + } + + return sched; +} + +void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler) +{ + if (scheduler->module) + module_put(scheduler->module); +} + + +/* + * Register a scheduler in the scheduler list + */ +int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) +{ + struct ip_vs_scheduler *sched; + + if (!scheduler) { + IP_VS_ERR("register_ip_vs_scheduler(): NULL arg\n"); + return -EINVAL; + } + + if (!scheduler->name) { + IP_VS_ERR("register_ip_vs_scheduler(): NULL scheduler_name\n"); + return -EINVAL; + } + + /* increase the module use count */ + ip_vs_use_count_inc(); + + write_lock_bh(&__ip_vs_sched_lock); + + if (!list_empty(&scheduler->n_list)) { + write_unlock_bh(&__ip_vs_sched_lock); + ip_vs_use_count_dec(); + IP_VS_ERR("register_ip_vs_scheduler(): [%s] scheduler " + "already linked\n", scheduler->name); + return -EINVAL; + } + + /* + * Make sure that the scheduler with this name doesn't exist + * in the scheduler list. + */ + list_for_each_entry(sched, &ip_vs_schedulers, n_list) { + if (strcmp(scheduler->name, sched->name) == 0) { + write_unlock_bh(&__ip_vs_sched_lock); + ip_vs_use_count_dec(); + IP_VS_ERR("register_ip_vs_scheduler(): [%s] scheduler " + "already existed in the system\n", + scheduler->name); + return -EINVAL; + } + } + /* + * Add it into the d-linked scheduler list + */ + list_add(&scheduler->n_list, &ip_vs_schedulers); + write_unlock_bh(&__ip_vs_sched_lock); + + IP_VS_INFO("[%s] scheduler registered.\n", scheduler->name); + + return 0; +} + + +/* + * Unregister a scheduler from the scheduler list + */ +int unregister_ip_vs_scheduler(struct ip_vs_scheduler *scheduler) +{ + if (!scheduler) { + IP_VS_ERR( "unregister_ip_vs_scheduler(): NULL arg\n"); + return -EINVAL; + } + + write_lock_bh(&__ip_vs_sched_lock); + if (list_empty(&scheduler->n_list)) { + write_unlock_bh(&__ip_vs_sched_lock); + IP_VS_ERR("unregister_ip_vs_scheduler(): [%s] scheduler " + "is not in the list. failed\n", scheduler->name); + return -EINVAL; + } + + /* + * Remove it from the d-linked scheduler list + */ + list_del(&scheduler->n_list); + write_unlock_bh(&__ip_vs_sched_lock); + + /* decrease the module use count */ + ip_vs_use_count_dec(); + + IP_VS_INFO("[%s] scheduler unregistered.\n", scheduler->name); + + return 0; +} diff --git a/net/netfilter/ipvs/ip_vs_sed.c b/net/netfilter/ipvs/ip_vs_sed.c new file mode 100644 index 00000000000..7d2f22f04b8 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_sed.c @@ -0,0 +1,140 @@ +/* + * IPVS: Shortest Expected Delay scheduling module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +/* + * The SED algorithm attempts to minimize each job's expected delay until + * completion. The expected delay that the job will experience is + * (Ci + 1) / Ui if sent to the ith server, in which Ci is the number of + * jobs on the ith server and Ui is the fixed service rate (weight) of + * the ith server. The SED algorithm adopts a greedy policy that each does + * what is in its own best interest, i.e. to join the queue which would + * minimize its expected delay of completion. + * + * See the following paper for more information: + * A. Weinrib and S. Shenker, Greed is not enough: Adaptive load sharing + * in large heterogeneous systems. In Proceedings IEEE INFOCOM'88, + * pages 986-994, 1988. + * + * Thanks must go to Marko Buuri <marko@buuri.name> for talking SED to me. + * + * The difference between SED and WLC is that SED includes the incoming + * job in the cost function (the increment of 1). SED may outperform + * WLC, while scheduling big jobs under larger heterogeneous systems + * (the server weight varies a lot). + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> + +#include <net/ip_vs.h> + + +static inline unsigned int +ip_vs_sed_dest_overhead(struct ip_vs_dest *dest) +{ + /* + * We only use the active connection number in the cost + * calculation here. + */ + return atomic_read(&dest->activeconns) + 1; +} + + +/* + * Weighted Least Connection scheduling + */ +static struct ip_vs_dest * +ip_vs_sed_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest, *least; + unsigned int loh, doh; + + IP_VS_DBG(6, "ip_vs_sed_schedule(): Scheduling...\n"); + + /* + * We calculate the load of each dest server as follows: + * (server expected overhead) / dest->weight + * + * Remember -- no floats in kernel mode!!! + * The comparison of h1*w2 > h2*w1 is equivalent to that of + * h1/w1 > h2/w2 + * if every weight is larger than zero. + * + * The server with weight=0 is quiesced and will not receive any + * new connections. + */ + + list_for_each_entry(dest, &svc->destinations, n_list) { + if (!(dest->flags & IP_VS_DEST_F_OVERLOAD) && + atomic_read(&dest->weight) > 0) { + least = dest; + loh = ip_vs_sed_dest_overhead(least); + goto nextstage; + } + } + return NULL; + + /* + * Find the destination with the least load. + */ + nextstage: + list_for_each_entry_continue(dest, &svc->destinations, n_list) { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + doh = ip_vs_sed_dest_overhead(dest); + if (loh * atomic_read(&dest->weight) > + doh * atomic_read(&least->weight)) { + least = dest; + loh = doh; + } + } + + IP_VS_DBG_BUF(6, "SED: server %s:%u " + "activeconns %d refcnt %d weight %d overhead %d\n", + IP_VS_DBG_ADDR(svc->af, &least->addr), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->refcnt), + atomic_read(&least->weight), loh); + + return least; +} + + +static struct ip_vs_scheduler ip_vs_sed_scheduler = +{ + .name = "sed", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_sed_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 1, +#endif + .schedule = ip_vs_sed_schedule, +}; + + +static int __init ip_vs_sed_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_sed_scheduler); +} + +static void __exit ip_vs_sed_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_sed_scheduler); +} + +module_init(ip_vs_sed_init); +module_exit(ip_vs_sed_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_sh.c b/net/netfilter/ipvs/ip_vs_sh.c new file mode 100644 index 00000000000..1d96de27fef --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_sh.c @@ -0,0 +1,258 @@ +/* + * IPVS: Source Hashing scheduling module + * + * Authors: Wensong Zhang <wensong@gnuchina.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +/* + * The sh algorithm is to select server by the hash key of source IP + * address. The pseudo code is as follows: + * + * n <- servernode[src_ip]; + * if (n is dead) OR + * (n is overloaded) or (n.weight <= 0) then + * return NULL; + * + * return n; + * + * Notes that servernode is a 256-bucket hash table that maps the hash + * index derived from packet source IP address to the current server + * array. If the sh scheduler is used in cache cluster, it is good to + * combine it with cache_bypass feature. When the statically assigned + * server is dead or overloaded, the load balancer can bypass the cache + * server and send requests to the original server directly. + * + */ + +#include <linux/ip.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> + +#include <net/ip_vs.h> + + +/* + * IPVS SH bucket + */ +struct ip_vs_sh_bucket { + struct ip_vs_dest *dest; /* real server (cache) */ +}; + +/* + * for IPVS SH entry hash table + */ +#ifndef CONFIG_IP_VS_SH_TAB_BITS +#define CONFIG_IP_VS_SH_TAB_BITS 8 +#endif +#define IP_VS_SH_TAB_BITS CONFIG_IP_VS_SH_TAB_BITS +#define IP_VS_SH_TAB_SIZE (1 << IP_VS_SH_TAB_BITS) +#define IP_VS_SH_TAB_MASK (IP_VS_SH_TAB_SIZE - 1) + + +/* + * Returns hash value for IPVS SH entry + */ +static inline unsigned ip_vs_sh_hashkey(__be32 addr) +{ + return (ntohl(addr)*2654435761UL) & IP_VS_SH_TAB_MASK; +} + + +/* + * Get ip_vs_dest associated with supplied parameters. + */ +static inline struct ip_vs_dest * +ip_vs_sh_get(struct ip_vs_sh_bucket *tbl, __be32 addr) +{ + return (tbl[ip_vs_sh_hashkey(addr)]).dest; +} + + +/* + * Assign all the hash buckets of the specified table with the service. + */ +static int +ip_vs_sh_assign(struct ip_vs_sh_bucket *tbl, struct ip_vs_service *svc) +{ + int i; + struct ip_vs_sh_bucket *b; + struct list_head *p; + struct ip_vs_dest *dest; + + b = tbl; + p = &svc->destinations; + for (i=0; i<IP_VS_SH_TAB_SIZE; i++) { + if (list_empty(p)) { + b->dest = NULL; + } else { + if (p == &svc->destinations) + p = p->next; + + dest = list_entry(p, struct ip_vs_dest, n_list); + atomic_inc(&dest->refcnt); + b->dest = dest; + + p = p->next; + } + b++; + } + return 0; +} + + +/* + * Flush all the hash buckets of the specified table. + */ +static void ip_vs_sh_flush(struct ip_vs_sh_bucket *tbl) +{ + int i; + struct ip_vs_sh_bucket *b; + + b = tbl; + for (i=0; i<IP_VS_SH_TAB_SIZE; i++) { + if (b->dest) { + atomic_dec(&b->dest->refcnt); + b->dest = NULL; + } + b++; + } +} + + +static int ip_vs_sh_init_svc(struct ip_vs_service *svc) +{ + struct ip_vs_sh_bucket *tbl; + + /* allocate the SH table for this service */ + tbl = kmalloc(sizeof(struct ip_vs_sh_bucket)*IP_VS_SH_TAB_SIZE, + GFP_ATOMIC); + if (tbl == NULL) { + IP_VS_ERR("ip_vs_sh_init_svc(): no memory\n"); + return -ENOMEM; + } + svc->sched_data = tbl; + IP_VS_DBG(6, "SH hash table (memory=%Zdbytes) allocated for " + "current service\n", + sizeof(struct ip_vs_sh_bucket)*IP_VS_SH_TAB_SIZE); + + /* assign the hash buckets with the updated service */ + ip_vs_sh_assign(tbl, svc); + + return 0; +} + + +static int ip_vs_sh_done_svc(struct ip_vs_service *svc) +{ + struct ip_vs_sh_bucket *tbl = svc->sched_data; + + /* got to clean up hash buckets here */ + ip_vs_sh_flush(tbl); + + /* release the table itself */ + kfree(svc->sched_data); + IP_VS_DBG(6, "SH hash table (memory=%Zdbytes) released\n", + sizeof(struct ip_vs_sh_bucket)*IP_VS_SH_TAB_SIZE); + + return 0; +} + + +static int ip_vs_sh_update_svc(struct ip_vs_service *svc) +{ + struct ip_vs_sh_bucket *tbl = svc->sched_data; + + /* got to clean up hash buckets here */ + ip_vs_sh_flush(tbl); + + /* assign the hash buckets with the updated service */ + ip_vs_sh_assign(tbl, svc); + + return 0; +} + + +/* + * If the dest flags is set with IP_VS_DEST_F_OVERLOAD, + * consider that the server is overloaded here. + */ +static inline int is_overloaded(struct ip_vs_dest *dest) +{ + return dest->flags & IP_VS_DEST_F_OVERLOAD; +} + + +/* + * Source Hashing scheduling + */ +static struct ip_vs_dest * +ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest; + struct ip_vs_sh_bucket *tbl; + struct iphdr *iph = ip_hdr(skb); + + IP_VS_DBG(6, "ip_vs_sh_schedule(): Scheduling...\n"); + + tbl = (struct ip_vs_sh_bucket *)svc->sched_data; + dest = ip_vs_sh_get(tbl, iph->saddr); + if (!dest + || !(dest->flags & IP_VS_DEST_F_AVAILABLE) + || atomic_read(&dest->weight) <= 0 + || is_overloaded(dest)) { + return NULL; + } + + IP_VS_DBG(6, "SH: source IP address %u.%u.%u.%u " + "--> server %u.%u.%u.%u:%d\n", + NIPQUAD(iph->saddr), + NIPQUAD(dest->addr.ip), + ntohs(dest->port)); + + return dest; +} + + +/* + * IPVS SH Scheduler structure + */ +static struct ip_vs_scheduler ip_vs_sh_scheduler = +{ + .name = "sh", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_sh_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 0, +#endif + .init_service = ip_vs_sh_init_svc, + .done_service = ip_vs_sh_done_svc, + .update_service = ip_vs_sh_update_svc, + .schedule = ip_vs_sh_schedule, +}; + + +static int __init ip_vs_sh_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_sh_scheduler); +} + + +static void __exit ip_vs_sh_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_sh_scheduler); +} + + +module_init(ip_vs_sh_init); +module_exit(ip_vs_sh_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c new file mode 100644 index 00000000000..de5e7e118ee --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -0,0 +1,942 @@ +/* + * IPVS An implementation of the IP virtual server support for the + * LINUX operating system. IPVS is now implemented as a module + * over the NetFilter framework. IPVS can be used to build a + * high-performance and highly available server based on a + * cluster of servers. + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * ip_vs_sync: sync connection info from master load balancer to backups + * through multicast + * + * Changes: + * Alexandre Cassen : Added master & backup support at a time. + * Alexandre Cassen : Added SyncID support for incoming sync + * messages filtering. + * Justin Ossevoort : Fix endian problem on sync message size. + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/inetdevice.h> +#include <linux/net.h> +#include <linux/completion.h> +#include <linux/delay.h> +#include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/igmp.h> /* for ip_mc_join_group */ +#include <linux/udp.h> +#include <linux/err.h> +#include <linux/kthread.h> +#include <linux/wait.h> +#include <linux/kernel.h> + +#include <net/ip.h> +#include <net/sock.h> + +#include <net/ip_vs.h> + +#define IP_VS_SYNC_GROUP 0xe0000051 /* multicast addr - 224.0.0.81 */ +#define IP_VS_SYNC_PORT 8848 /* multicast port */ + + +/* + * IPVS sync connection entry + */ +struct ip_vs_sync_conn { + __u8 reserved; + + /* Protocol, addresses and port numbers */ + __u8 protocol; /* Which protocol (TCP/UDP) */ + __be16 cport; + __be16 vport; + __be16 dport; + __be32 caddr; /* client address */ + __be32 vaddr; /* virtual address */ + __be32 daddr; /* destination address */ + + /* Flags and state transition */ + __be16 flags; /* status flags */ + __be16 state; /* state info */ + + /* The sequence options start here */ +}; + +struct ip_vs_sync_conn_options { + struct ip_vs_seq in_seq; /* incoming seq. struct */ + struct ip_vs_seq out_seq; /* outgoing seq. struct */ +}; + +struct ip_vs_sync_thread_data { + struct socket *sock; + char *buf; +}; + +#define SIMPLE_CONN_SIZE (sizeof(struct ip_vs_sync_conn)) +#define FULL_CONN_SIZE \ +(sizeof(struct ip_vs_sync_conn) + sizeof(struct ip_vs_sync_conn_options)) + + +/* + The master mulitcasts messages to the backup load balancers in the + following format. + + 0 1 2 3 + 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Count Conns | SyncID | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | | + | IPVS Sync Connection (1) | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | . | + | . | + | . | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | | + | IPVS Sync Connection (n) | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +*/ + +#define SYNC_MESG_HEADER_LEN 4 +#define MAX_CONNS_PER_SYNCBUFF 255 /* nr_conns in ip_vs_sync_mesg is 8 bit */ + +struct ip_vs_sync_mesg { + __u8 nr_conns; + __u8 syncid; + __u16 size; + + /* ip_vs_sync_conn entries start here */ +}; + +/* the maximum length of sync (sending/receiving) message */ +static int sync_send_mesg_maxlen; +static int sync_recv_mesg_maxlen; + +struct ip_vs_sync_buff { + struct list_head list; + unsigned long firstuse; + + /* pointers for the message data */ + struct ip_vs_sync_mesg *mesg; + unsigned char *head; + unsigned char *end; +}; + + +/* the sync_buff list head and the lock */ +static LIST_HEAD(ip_vs_sync_queue); +static DEFINE_SPINLOCK(ip_vs_sync_lock); + +/* current sync_buff for accepting new conn entries */ +static struct ip_vs_sync_buff *curr_sb = NULL; +static DEFINE_SPINLOCK(curr_sb_lock); + +/* ipvs sync daemon state */ +volatile int ip_vs_sync_state = IP_VS_STATE_NONE; +volatile int ip_vs_master_syncid = 0; +volatile int ip_vs_backup_syncid = 0; + +/* multicast interface name */ +char ip_vs_master_mcast_ifn[IP_VS_IFNAME_MAXLEN]; +char ip_vs_backup_mcast_ifn[IP_VS_IFNAME_MAXLEN]; + +/* sync daemon tasks */ +static struct task_struct *sync_master_thread; +static struct task_struct *sync_backup_thread; + +/* multicast addr */ +static struct sockaddr_in mcast_addr = { + .sin_family = AF_INET, + .sin_port = __constant_htons(IP_VS_SYNC_PORT), + .sin_addr.s_addr = __constant_htonl(IP_VS_SYNC_GROUP), +}; + + +static inline struct ip_vs_sync_buff *sb_dequeue(void) +{ + struct ip_vs_sync_buff *sb; + + spin_lock_bh(&ip_vs_sync_lock); + if (list_empty(&ip_vs_sync_queue)) { + sb = NULL; + } else { + sb = list_entry(ip_vs_sync_queue.next, + struct ip_vs_sync_buff, + list); + list_del(&sb->list); + } + spin_unlock_bh(&ip_vs_sync_lock); + + return sb; +} + +static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create(void) +{ + struct ip_vs_sync_buff *sb; + + if (!(sb=kmalloc(sizeof(struct ip_vs_sync_buff), GFP_ATOMIC))) + return NULL; + + if (!(sb->mesg=kmalloc(sync_send_mesg_maxlen, GFP_ATOMIC))) { + kfree(sb); + return NULL; + } + sb->mesg->nr_conns = 0; + sb->mesg->syncid = ip_vs_master_syncid; + sb->mesg->size = 4; + sb->head = (unsigned char *)sb->mesg + 4; + sb->end = (unsigned char *)sb->mesg + sync_send_mesg_maxlen; + sb->firstuse = jiffies; + return sb; +} + +static inline void ip_vs_sync_buff_release(struct ip_vs_sync_buff *sb) +{ + kfree(sb->mesg); + kfree(sb); +} + +static inline void sb_queue_tail(struct ip_vs_sync_buff *sb) +{ + spin_lock(&ip_vs_sync_lock); + if (ip_vs_sync_state & IP_VS_STATE_MASTER) + list_add_tail(&sb->list, &ip_vs_sync_queue); + else + ip_vs_sync_buff_release(sb); + spin_unlock(&ip_vs_sync_lock); +} + +/* + * Get the current sync buffer if it has been created for more + * than the specified time or the specified time is zero. + */ +static inline struct ip_vs_sync_buff * +get_curr_sync_buff(unsigned long time) +{ + struct ip_vs_sync_buff *sb; + + spin_lock_bh(&curr_sb_lock); + if (curr_sb && (time == 0 || + time_before(jiffies - curr_sb->firstuse, time))) { + sb = curr_sb; + curr_sb = NULL; + } else + sb = NULL; + spin_unlock_bh(&curr_sb_lock); + return sb; +} + + +/* + * Add an ip_vs_conn information into the current sync_buff. + * Called by ip_vs_in. + */ +void ip_vs_sync_conn(struct ip_vs_conn *cp) +{ + struct ip_vs_sync_mesg *m; + struct ip_vs_sync_conn *s; + int len; + + spin_lock(&curr_sb_lock); + if (!curr_sb) { + if (!(curr_sb=ip_vs_sync_buff_create())) { + spin_unlock(&curr_sb_lock); + IP_VS_ERR("ip_vs_sync_buff_create failed.\n"); + return; + } + } + + len = (cp->flags & IP_VS_CONN_F_SEQ_MASK) ? FULL_CONN_SIZE : + SIMPLE_CONN_SIZE; + m = curr_sb->mesg; + s = (struct ip_vs_sync_conn *)curr_sb->head; + + /* copy members */ + s->protocol = cp->protocol; + s->cport = cp->cport; + s->vport = cp->vport; + s->dport = cp->dport; + s->caddr = cp->caddr.ip; + s->vaddr = cp->vaddr.ip; + s->daddr = cp->daddr.ip; + s->flags = htons(cp->flags & ~IP_VS_CONN_F_HASHED); + s->state = htons(cp->state); + if (cp->flags & IP_VS_CONN_F_SEQ_MASK) { + struct ip_vs_sync_conn_options *opt = + (struct ip_vs_sync_conn_options *)&s[1]; + memcpy(opt, &cp->in_seq, sizeof(*opt)); + } + + m->nr_conns++; + m->size += len; + curr_sb->head += len; + + /* check if there is a space for next one */ + if (curr_sb->head+FULL_CONN_SIZE > curr_sb->end) { + sb_queue_tail(curr_sb); + curr_sb = NULL; + } + spin_unlock(&curr_sb_lock); + + /* synchronize its controller if it has */ + if (cp->control) + ip_vs_sync_conn(cp->control); +} + + +/* + * Process received multicast message and create the corresponding + * ip_vs_conn entries. + */ +static void ip_vs_process_message(const char *buffer, const size_t buflen) +{ + struct ip_vs_sync_mesg *m = (struct ip_vs_sync_mesg *)buffer; + struct ip_vs_sync_conn *s; + struct ip_vs_sync_conn_options *opt; + struct ip_vs_conn *cp; + struct ip_vs_protocol *pp; + struct ip_vs_dest *dest; + char *p; + int i; + + if (buflen < sizeof(struct ip_vs_sync_mesg)) { + IP_VS_ERR_RL("sync message header too short\n"); + return; + } + + /* Convert size back to host byte order */ + m->size = ntohs(m->size); + + if (buflen != m->size) { + IP_VS_ERR_RL("bogus sync message size\n"); + return; + } + + /* SyncID sanity check */ + if (ip_vs_backup_syncid != 0 && m->syncid != ip_vs_backup_syncid) { + IP_VS_DBG(7, "Ignoring incoming msg with syncid = %d\n", + m->syncid); + return; + } + + p = (char *)buffer + sizeof(struct ip_vs_sync_mesg); + for (i=0; i<m->nr_conns; i++) { + unsigned flags, state; + + if (p + SIMPLE_CONN_SIZE > buffer+buflen) { + IP_VS_ERR_RL("bogus conn in sync message\n"); + return; + } + s = (struct ip_vs_sync_conn *) p; + flags = ntohs(s->flags) | IP_VS_CONN_F_SYNC; + flags &= ~IP_VS_CONN_F_HASHED; + if (flags & IP_VS_CONN_F_SEQ_MASK) { + opt = (struct ip_vs_sync_conn_options *)&s[1]; + p += FULL_CONN_SIZE; + if (p > buffer+buflen) { + IP_VS_ERR_RL("bogus conn options in sync message\n"); + return; + } + } else { + opt = NULL; + p += SIMPLE_CONN_SIZE; + } + + state = ntohs(s->state); + if (!(flags & IP_VS_CONN_F_TEMPLATE)) { + pp = ip_vs_proto_get(s->protocol); + if (!pp) { + IP_VS_ERR_RL("Unsupported protocol %u in sync msg\n", + s->protocol); + continue; + } + if (state >= pp->num_states) { + IP_VS_DBG(2, "Invalid %s state %u in sync msg\n", + pp->name, state); + continue; + } + } else { + /* protocol in templates is not used for state/timeout */ + pp = NULL; + if (state > 0) { + IP_VS_DBG(2, "Invalid template state %u in sync msg\n", + state); + state = 0; + } + } + + if (!(flags & IP_VS_CONN_F_TEMPLATE)) + cp = ip_vs_conn_in_get(AF_INET, s->protocol, + (union nf_inet_addr *)&s->caddr, + s->cport, + (union nf_inet_addr *)&s->vaddr, + s->vport); + else + cp = ip_vs_ct_in_get(AF_INET, s->protocol, + (union nf_inet_addr *)&s->caddr, + s->cport, + (union nf_inet_addr *)&s->vaddr, + s->vport); + if (!cp) { + /* + * Find the appropriate destination for the connection. + * If it is not found the connection will remain unbound + * but still handled. + */ + dest = ip_vs_find_dest(AF_INET, + (union nf_inet_addr *)&s->daddr, + s->dport, + (union nf_inet_addr *)&s->vaddr, + s->vport, + s->protocol); + /* Set the approprite ativity flag */ + if (s->protocol == IPPROTO_TCP) { + if (state != IP_VS_TCP_S_ESTABLISHED) + flags |= IP_VS_CONN_F_INACTIVE; + else + flags &= ~IP_VS_CONN_F_INACTIVE; + } + cp = ip_vs_conn_new(AF_INET, s->protocol, + (union nf_inet_addr *)&s->caddr, + s->cport, + (union nf_inet_addr *)&s->vaddr, + s->vport, + (union nf_inet_addr *)&s->daddr, + s->dport, + flags, dest); + if (dest) + atomic_dec(&dest->refcnt); + if (!cp) { + IP_VS_ERR("ip_vs_conn_new failed\n"); + return; + } + } else if (!cp->dest) { + dest = ip_vs_try_bind_dest(cp); + if (dest) + atomic_dec(&dest->refcnt); + } else if ((cp->dest) && (cp->protocol == IPPROTO_TCP) && + (cp->state != state)) { + /* update active/inactive flag for the connection */ + dest = cp->dest; + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (state != IP_VS_TCP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags |= IP_VS_CONN_F_INACTIVE; + } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && + (state == IP_VS_TCP_S_ESTABLISHED)) { + atomic_inc(&dest->activeconns); + atomic_dec(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } + } + + if (opt) + memcpy(&cp->in_seq, opt, sizeof(*opt)); + atomic_set(&cp->in_pkts, sysctl_ip_vs_sync_threshold[0]); + cp->state = state; + cp->old_state = cp->state; + /* + * We can not recover the right timeout for templates + * in all cases, we can not find the right fwmark + * virtual service. If needed, we can do it for + * non-fwmark persistent services. + */ + if (!(flags & IP_VS_CONN_F_TEMPLATE) && pp->timeout_table) + cp->timeout = pp->timeout_table[state]; + else + cp->timeout = (3*60*HZ); + ip_vs_conn_put(cp); + } +} + + +/* + * Setup loopback of outgoing multicasts on a sending socket + */ +static void set_mcast_loop(struct sock *sk, u_char loop) +{ + struct inet_sock *inet = inet_sk(sk); + + /* setsockopt(sock, SOL_IP, IP_MULTICAST_LOOP, &loop, sizeof(loop)); */ + lock_sock(sk); + inet->mc_loop = loop ? 1 : 0; + release_sock(sk); +} + +/* + * Specify TTL for outgoing multicasts on a sending socket + */ +static void set_mcast_ttl(struct sock *sk, u_char ttl) +{ + struct inet_sock *inet = inet_sk(sk); + + /* setsockopt(sock, SOL_IP, IP_MULTICAST_TTL, &ttl, sizeof(ttl)); */ + lock_sock(sk); + inet->mc_ttl = ttl; + release_sock(sk); +} + +/* + * Specifiy default interface for outgoing multicasts + */ +static int set_mcast_if(struct sock *sk, char *ifname) +{ + struct net_device *dev; + struct inet_sock *inet = inet_sk(sk); + + if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + return -ENODEV; + + if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if) + return -EINVAL; + + lock_sock(sk); + inet->mc_index = dev->ifindex; + /* inet->mc_addr = 0; */ + release_sock(sk); + + return 0; +} + + +/* + * Set the maximum length of sync message according to the + * specified interface's MTU. + */ +static int set_sync_mesg_maxlen(int sync_state) +{ + struct net_device *dev; + int num; + + if (sync_state == IP_VS_STATE_MASTER) { + if ((dev = __dev_get_by_name(&init_net, ip_vs_master_mcast_ifn)) == NULL) + return -ENODEV; + + num = (dev->mtu - sizeof(struct iphdr) - + sizeof(struct udphdr) - + SYNC_MESG_HEADER_LEN - 20) / SIMPLE_CONN_SIZE; + sync_send_mesg_maxlen = SYNC_MESG_HEADER_LEN + + SIMPLE_CONN_SIZE * min(num, MAX_CONNS_PER_SYNCBUFF); + IP_VS_DBG(7, "setting the maximum length of sync sending " + "message %d.\n", sync_send_mesg_maxlen); + } else if (sync_state == IP_VS_STATE_BACKUP) { + if ((dev = __dev_get_by_name(&init_net, ip_vs_backup_mcast_ifn)) == NULL) + return -ENODEV; + + sync_recv_mesg_maxlen = dev->mtu - + sizeof(struct iphdr) - sizeof(struct udphdr); + IP_VS_DBG(7, "setting the maximum length of sync receiving " + "message %d.\n", sync_recv_mesg_maxlen); + } + + return 0; +} + + +/* + * Join a multicast group. + * the group is specified by a class D multicast address 224.0.0.0/8 + * in the in_addr structure passed in as a parameter. + */ +static int +join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) +{ + struct ip_mreqn mreq; + struct net_device *dev; + int ret; + + memset(&mreq, 0, sizeof(mreq)); + memcpy(&mreq.imr_multiaddr, addr, sizeof(struct in_addr)); + + if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + return -ENODEV; + if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if) + return -EINVAL; + + mreq.imr_ifindex = dev->ifindex; + + lock_sock(sk); + ret = ip_mc_join_group(sk, &mreq); + release_sock(sk); + + return ret; +} + + +static int bind_mcastif_addr(struct socket *sock, char *ifname) +{ + struct net_device *dev; + __be32 addr; + struct sockaddr_in sin; + + if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + return -ENODEV; + + addr = inet_select_addr(dev, 0, RT_SCOPE_UNIVERSE); + if (!addr) + IP_VS_ERR("You probably need to specify IP address on " + "multicast interface.\n"); + + IP_VS_DBG(7, "binding socket with (%s) %u.%u.%u.%u\n", + ifname, NIPQUAD(addr)); + + /* Now bind the socket with the address of multicast interface */ + sin.sin_family = AF_INET; + sin.sin_addr.s_addr = addr; + sin.sin_port = 0; + + return sock->ops->bind(sock, (struct sockaddr*)&sin, sizeof(sin)); +} + +/* + * Set up sending multicast socket over UDP + */ +static struct socket * make_send_sock(void) +{ + struct socket *sock; + int result; + + /* First create a socket */ + result = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + if (result < 0) { + IP_VS_ERR("Error during creation of socket; terminating\n"); + return ERR_PTR(result); + } + + result = set_mcast_if(sock->sk, ip_vs_master_mcast_ifn); + if (result < 0) { + IP_VS_ERR("Error setting outbound mcast interface\n"); + goto error; + } + + set_mcast_loop(sock->sk, 0); + set_mcast_ttl(sock->sk, 1); + + result = bind_mcastif_addr(sock, ip_vs_master_mcast_ifn); + if (result < 0) { + IP_VS_ERR("Error binding address of the mcast interface\n"); + goto error; + } + + result = sock->ops->connect(sock, (struct sockaddr *) &mcast_addr, + sizeof(struct sockaddr), 0); + if (result < 0) { + IP_VS_ERR("Error connecting to the multicast addr\n"); + goto error; + } + + return sock; + + error: + sock_release(sock); + return ERR_PTR(result); +} + + +/* + * Set up receiving multicast socket over UDP + */ +static struct socket * make_receive_sock(void) +{ + struct socket *sock; + int result; + + /* First create a socket */ + result = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + if (result < 0) { + IP_VS_ERR("Error during creation of socket; terminating\n"); + return ERR_PTR(result); + } + + /* it is equivalent to the REUSEADDR option in user-space */ + sock->sk->sk_reuse = 1; + + result = sock->ops->bind(sock, (struct sockaddr *) &mcast_addr, + sizeof(struct sockaddr)); + if (result < 0) { + IP_VS_ERR("Error binding to the multicast addr\n"); + goto error; + } + + /* join the multicast group */ + result = join_mcast_group(sock->sk, + (struct in_addr *) &mcast_addr.sin_addr, + ip_vs_backup_mcast_ifn); + if (result < 0) { + IP_VS_ERR("Error joining to the multicast group\n"); + goto error; + } + + return sock; + + error: + sock_release(sock); + return ERR_PTR(result); +} + + +static int +ip_vs_send_async(struct socket *sock, const char *buffer, const size_t length) +{ + struct msghdr msg = {.msg_flags = MSG_DONTWAIT|MSG_NOSIGNAL}; + struct kvec iov; + int len; + + EnterFunction(7); + iov.iov_base = (void *)buffer; + iov.iov_len = length; + + len = kernel_sendmsg(sock, &msg, &iov, 1, (size_t)(length)); + + LeaveFunction(7); + return len; +} + +static void +ip_vs_send_sync_msg(struct socket *sock, struct ip_vs_sync_mesg *msg) +{ + int msize; + + msize = msg->size; + + /* Put size in network byte order */ + msg->size = htons(msg->size); + + if (ip_vs_send_async(sock, (char *)msg, msize) != msize) + IP_VS_ERR("ip_vs_send_async error\n"); +} + +static int +ip_vs_receive(struct socket *sock, char *buffer, const size_t buflen) +{ + struct msghdr msg = {NULL,}; + struct kvec iov; + int len; + + EnterFunction(7); + + /* Receive a packet */ + iov.iov_base = buffer; + iov.iov_len = (size_t)buflen; + + len = kernel_recvmsg(sock, &msg, &iov, 1, buflen, 0); + + if (len < 0) + return -1; + + LeaveFunction(7); + return len; +} + + +static int sync_thread_master(void *data) +{ + struct ip_vs_sync_thread_data *tinfo = data; + struct ip_vs_sync_buff *sb; + + IP_VS_INFO("sync thread started: state = MASTER, mcast_ifn = %s, " + "syncid = %d\n", + ip_vs_master_mcast_ifn, ip_vs_master_syncid); + + while (!kthread_should_stop()) { + while ((sb = sb_dequeue())) { + ip_vs_send_sync_msg(tinfo->sock, sb->mesg); + ip_vs_sync_buff_release(sb); + } + + /* check if entries stay in curr_sb for 2 seconds */ + sb = get_curr_sync_buff(2 * HZ); + if (sb) { + ip_vs_send_sync_msg(tinfo->sock, sb->mesg); + ip_vs_sync_buff_release(sb); + } + + schedule_timeout_interruptible(HZ); + } + + /* clean up the sync_buff queue */ + while ((sb=sb_dequeue())) { + ip_vs_sync_buff_release(sb); + } + + /* clean up the current sync_buff */ + if ((sb = get_curr_sync_buff(0))) { + ip_vs_sync_buff_release(sb); + } + + /* release the sending multicast socket */ + sock_release(tinfo->sock); + kfree(tinfo); + + return 0; +} + + +static int sync_thread_backup(void *data) +{ + struct ip_vs_sync_thread_data *tinfo = data; + int len; + + IP_VS_INFO("sync thread started: state = BACKUP, mcast_ifn = %s, " + "syncid = %d\n", + ip_vs_backup_mcast_ifn, ip_vs_backup_syncid); + + while (!kthread_should_stop()) { + wait_event_interruptible(*tinfo->sock->sk->sk_sleep, + !skb_queue_empty(&tinfo->sock->sk->sk_receive_queue) + || kthread_should_stop()); + + /* do we have data now? */ + while (!skb_queue_empty(&(tinfo->sock->sk->sk_receive_queue))) { + len = ip_vs_receive(tinfo->sock, tinfo->buf, + sync_recv_mesg_maxlen); + if (len <= 0) { + IP_VS_ERR("receiving message error\n"); + break; + } + + /* disable bottom half, because it accesses the data + shared by softirq while getting/creating conns */ + local_bh_disable(); + ip_vs_process_message(tinfo->buf, len); + local_bh_enable(); + } + } + + /* release the sending multicast socket */ + sock_release(tinfo->sock); + kfree(tinfo->buf); + kfree(tinfo); + + return 0; +} + + +int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) +{ + struct ip_vs_sync_thread_data *tinfo; + struct task_struct **realtask, *task; + struct socket *sock; + char *name, *buf = NULL; + int (*threadfn)(void *data); + int result = -ENOMEM; + + IP_VS_DBG(7, "%s: pid %d\n", __func__, task_pid_nr(current)); + IP_VS_DBG(7, "Each ip_vs_sync_conn entry needs %Zd bytes\n", + sizeof(struct ip_vs_sync_conn)); + + if (state == IP_VS_STATE_MASTER) { + if (sync_master_thread) + return -EEXIST; + + strlcpy(ip_vs_master_mcast_ifn, mcast_ifn, + sizeof(ip_vs_master_mcast_ifn)); + ip_vs_master_syncid = syncid; + realtask = &sync_master_thread; + name = "ipvs_syncmaster"; + threadfn = sync_thread_master; + sock = make_send_sock(); + } else if (state == IP_VS_STATE_BACKUP) { + if (sync_backup_thread) + return -EEXIST; + + strlcpy(ip_vs_backup_mcast_ifn, mcast_ifn, + sizeof(ip_vs_backup_mcast_ifn)); + ip_vs_backup_syncid = syncid; + realtask = &sync_backup_thread; + name = "ipvs_syncbackup"; + threadfn = sync_thread_backup; + sock = make_receive_sock(); + } else { + return -EINVAL; + } + + if (IS_ERR(sock)) { + result = PTR_ERR(sock); + goto out; + } + + set_sync_mesg_maxlen(state); + if (state == IP_VS_STATE_BACKUP) { + buf = kmalloc(sync_recv_mesg_maxlen, GFP_KERNEL); + if (!buf) + goto outsocket; + } + + tinfo = kmalloc(sizeof(*tinfo), GFP_KERNEL); + if (!tinfo) + goto outbuf; + + tinfo->sock = sock; + tinfo->buf = buf; + + task = kthread_run(threadfn, tinfo, name); + if (IS_ERR(task)) { + result = PTR_ERR(task); + goto outtinfo; + } + + /* mark as active */ + *realtask = task; + ip_vs_sync_state |= state; + + /* increase the module use count */ + ip_vs_use_count_inc(); + + return 0; + +outtinfo: + kfree(tinfo); +outbuf: + kfree(buf); +outsocket: + sock_release(sock); +out: + return result; +} + + +int stop_sync_thread(int state) +{ + IP_VS_DBG(7, "%s: pid %d\n", __func__, task_pid_nr(current)); + + if (state == IP_VS_STATE_MASTER) { + if (!sync_master_thread) + return -ESRCH; + + IP_VS_INFO("stopping master sync thread %d ...\n", + task_pid_nr(sync_master_thread)); + + /* + * The lock synchronizes with sb_queue_tail(), so that we don't + * add sync buffers to the queue, when we are already in + * progress of stopping the master sync daemon. + */ + + spin_lock_bh(&ip_vs_sync_lock); + ip_vs_sync_state &= ~IP_VS_STATE_MASTER; + spin_unlock_bh(&ip_vs_sync_lock); + kthread_stop(sync_master_thread); + sync_master_thread = NULL; + } else if (state == IP_VS_STATE_BACKUP) { + if (!sync_backup_thread) + return -ESRCH; + + IP_VS_INFO("stopping backup sync thread %d ...\n", + task_pid_nr(sync_backup_thread)); + + ip_vs_sync_state &= ~IP_VS_STATE_BACKUP; + kthread_stop(sync_backup_thread); + sync_backup_thread = NULL; + } else { + return -EINVAL; + } + + /* decrease the module use count */ + ip_vs_use_count_dec(); + + return 0; +} diff --git a/net/netfilter/ipvs/ip_vs_wlc.c b/net/netfilter/ipvs/ip_vs_wlc.c new file mode 100644 index 00000000000..8c596e71259 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_wlc.c @@ -0,0 +1,128 @@ +/* + * IPVS: Weighted Least-Connection Scheduling module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Peter Kese <peter.kese@ijs.si> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * Wensong Zhang : changed the ip_vs_wlc_schedule to return dest + * Wensong Zhang : changed to use the inactconns in scheduling + * Wensong Zhang : changed some comestics things for debugging + * Wensong Zhang : changed for the d-linked destination list + * Wensong Zhang : added the ip_vs_wlc_update_svc + * Wensong Zhang : added any dest with weight=0 is quiesced + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> + +#include <net/ip_vs.h> + + +static inline unsigned int +ip_vs_wlc_dest_overhead(struct ip_vs_dest *dest) +{ + /* + * We think the overhead of processing active connections is 256 + * times higher than that of inactive connections in average. (This + * 256 times might not be accurate, we will change it later) We + * use the following formula to estimate the overhead now: + * dest->activeconns*256 + dest->inactconns + */ + return (atomic_read(&dest->activeconns) << 8) + + atomic_read(&dest->inactconns); +} + + +/* + * Weighted Least Connection scheduling + */ +static struct ip_vs_dest * +ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest, *least; + unsigned int loh, doh; + + IP_VS_DBG(6, "ip_vs_wlc_schedule(): Scheduling...\n"); + + /* + * We calculate the load of each dest server as follows: + * (dest overhead) / dest->weight + * + * Remember -- no floats in kernel mode!!! + * The comparison of h1*w2 > h2*w1 is equivalent to that of + * h1/w1 > h2/w2 + * if every weight is larger than zero. + * + * The server with weight=0 is quiesced and will not receive any + * new connections. + */ + + list_for_each_entry(dest, &svc->destinations, n_list) { + if (!(dest->flags & IP_VS_DEST_F_OVERLOAD) && + atomic_read(&dest->weight) > 0) { + least = dest; + loh = ip_vs_wlc_dest_overhead(least); + goto nextstage; + } + } + return NULL; + + /* + * Find the destination with the least load. + */ + nextstage: + list_for_each_entry_continue(dest, &svc->destinations, n_list) { + if (dest->flags & IP_VS_DEST_F_OVERLOAD) + continue; + doh = ip_vs_wlc_dest_overhead(dest); + if (loh * atomic_read(&dest->weight) > + doh * atomic_read(&least->weight)) { + least = dest; + loh = doh; + } + } + + IP_VS_DBG_BUF(6, "WLC: server %s:%u " + "activeconns %d refcnt %d weight %d overhead %d\n", + IP_VS_DBG_ADDR(svc->af, &least->addr), ntohs(least->port), + atomic_read(&least->activeconns), + atomic_read(&least->refcnt), + atomic_read(&least->weight), loh); + + return least; +} + + +static struct ip_vs_scheduler ip_vs_wlc_scheduler = +{ + .name = "wlc", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_wlc_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 1, +#endif + .schedule = ip_vs_wlc_schedule, +}; + + +static int __init ip_vs_wlc_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_wlc_scheduler); +} + +static void __exit ip_vs_wlc_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_wlc_scheduler); +} + +module_init(ip_vs_wlc_init); +module_exit(ip_vs_wlc_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_wrr.c b/net/netfilter/ipvs/ip_vs_wrr.c new file mode 100644 index 00000000000..7ea92fed50b --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_wrr.c @@ -0,0 +1,237 @@ +/* + * IPVS: Weighted Round-Robin Scheduling module + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * Wensong Zhang : changed the ip_vs_wrr_schedule to return dest + * Wensong Zhang : changed some comestics things for debugging + * Wensong Zhang : changed for the d-linked destination list + * Wensong Zhang : added the ip_vs_wrr_update_svc + * Julian Anastasov : fixed the bug of returning destination + * with weight 0 when all weights are zero + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/net.h> + +#include <net/ip_vs.h> + +/* + * current destination pointer for weighted round-robin scheduling + */ +struct ip_vs_wrr_mark { + struct list_head *cl; /* current list head */ + int cw; /* current weight */ + int mw; /* maximum weight */ + int di; /* decreasing interval */ +}; + + +/* + * Get the gcd of server weights + */ +static int gcd(int a, int b) +{ + int c; + + while ((c = a % b)) { + a = b; + b = c; + } + return b; +} + +static int ip_vs_wrr_gcd_weight(struct ip_vs_service *svc) +{ + struct ip_vs_dest *dest; + int weight; + int g = 0; + + list_for_each_entry(dest, &svc->destinations, n_list) { + weight = atomic_read(&dest->weight); + if (weight > 0) { + if (g > 0) + g = gcd(weight, g); + else + g = weight; + } + } + return g ? g : 1; +} + + +/* + * Get the maximum weight of the service destinations. + */ +static int ip_vs_wrr_max_weight(struct ip_vs_service *svc) +{ + struct ip_vs_dest *dest; + int weight = 0; + + list_for_each_entry(dest, &svc->destinations, n_list) { + if (atomic_read(&dest->weight) > weight) + weight = atomic_read(&dest->weight); + } + + return weight; +} + + +static int ip_vs_wrr_init_svc(struct ip_vs_service *svc) +{ + struct ip_vs_wrr_mark *mark; + + /* + * Allocate the mark variable for WRR scheduling + */ + mark = kmalloc(sizeof(struct ip_vs_wrr_mark), GFP_ATOMIC); + if (mark == NULL) { + IP_VS_ERR("ip_vs_wrr_init_svc(): no memory\n"); + return -ENOMEM; + } + mark->cl = &svc->destinations; + mark->cw = 0; + mark->mw = ip_vs_wrr_max_weight(svc); + mark->di = ip_vs_wrr_gcd_weight(svc); + svc->sched_data = mark; + + return 0; +} + + +static int ip_vs_wrr_done_svc(struct ip_vs_service *svc) +{ + /* + * Release the mark variable + */ + kfree(svc->sched_data); + + return 0; +} + + +static int ip_vs_wrr_update_svc(struct ip_vs_service *svc) +{ + struct ip_vs_wrr_mark *mark = svc->sched_data; + + mark->cl = &svc->destinations; + mark->mw = ip_vs_wrr_max_weight(svc); + mark->di = ip_vs_wrr_gcd_weight(svc); + if (mark->cw > mark->mw) + mark->cw = 0; + return 0; +} + + +/* + * Weighted Round-Robin Scheduling + */ +static struct ip_vs_dest * +ip_vs_wrr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) +{ + struct ip_vs_dest *dest; + struct ip_vs_wrr_mark *mark = svc->sched_data; + struct list_head *p; + + IP_VS_DBG(6, "ip_vs_wrr_schedule(): Scheduling...\n"); + + /* + * This loop will always terminate, because mark->cw in (0, max_weight] + * and at least one server has its weight equal to max_weight. + */ + write_lock(&svc->sched_lock); + p = mark->cl; + while (1) { + if (mark->cl == &svc->destinations) { + /* it is at the head of the destination list */ + + if (mark->cl == mark->cl->next) { + /* no dest entry */ + dest = NULL; + goto out; + } + + mark->cl = svc->destinations.next; + mark->cw -= mark->di; + if (mark->cw <= 0) { + mark->cw = mark->mw; + /* + * Still zero, which means no available servers. + */ + if (mark->cw == 0) { + mark->cl = &svc->destinations; + IP_VS_ERR_RL("ip_vs_wrr_schedule(): " + "no available servers\n"); + dest = NULL; + goto out; + } + } + } else + mark->cl = mark->cl->next; + + if (mark->cl != &svc->destinations) { + /* not at the head of the list */ + dest = list_entry(mark->cl, struct ip_vs_dest, n_list); + if (!(dest->flags & IP_VS_DEST_F_OVERLOAD) && + atomic_read(&dest->weight) >= mark->cw) { + /* got it */ + break; + } + } + + if (mark->cl == p && mark->cw == mark->di) { + /* back to the start, and no dest is found. + It is only possible when all dests are OVERLOADED */ + dest = NULL; + goto out; + } + } + + IP_VS_DBG_BUF(6, "WRR: server %s:%u " + "activeconns %d refcnt %d weight %d\n", + IP_VS_DBG_ADDR(svc->af, &dest->addr), ntohs(dest->port), + atomic_read(&dest->activeconns), + atomic_read(&dest->refcnt), + atomic_read(&dest->weight)); + + out: + write_unlock(&svc->sched_lock); + return dest; +} + + +static struct ip_vs_scheduler ip_vs_wrr_scheduler = { + .name = "wrr", + .refcnt = ATOMIC_INIT(0), + .module = THIS_MODULE, + .n_list = LIST_HEAD_INIT(ip_vs_wrr_scheduler.n_list), +#ifdef CONFIG_IP_VS_IPV6 + .supports_ipv6 = 1, +#endif + .init_service = ip_vs_wrr_init_svc, + .done_service = ip_vs_wrr_done_svc, + .update_service = ip_vs_wrr_update_svc, + .schedule = ip_vs_wrr_schedule, +}; + +static int __init ip_vs_wrr_init(void) +{ + return register_ip_vs_scheduler(&ip_vs_wrr_scheduler) ; +} + +static void __exit ip_vs_wrr_cleanup(void) +{ + unregister_ip_vs_scheduler(&ip_vs_wrr_scheduler); +} + +module_init(ip_vs_wrr_init); +module_exit(ip_vs_wrr_cleanup); +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c new file mode 100644 index 00000000000..02ddc2b3ce2 --- /dev/null +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -0,0 +1,1004 @@ +/* + * ip_vs_xmit.c: various packet transmitters for IPVS + * + * Authors: Wensong Zhang <wensong@linuxvirtualserver.org> + * Julian Anastasov <ja@ssi.bg> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Changes: + * + */ + +#include <linux/kernel.h> +#include <linux/tcp.h> /* for tcphdr */ +#include <net/ip.h> +#include <net/tcp.h> /* for csum_tcpudp_magic */ +#include <net/udp.h> +#include <net/icmp.h> /* for icmp_send */ +#include <net/route.h> /* for ip_route_output */ +#include <net/ipv6.h> +#include <net/ip6_route.h> +#include <linux/icmpv6.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> + +#include <net/ip_vs.h> + + +/* + * Destination cache to speed up outgoing route lookup + */ +static inline void +__ip_vs_dst_set(struct ip_vs_dest *dest, u32 rtos, struct dst_entry *dst) +{ + struct dst_entry *old_dst; + + old_dst = dest->dst_cache; + dest->dst_cache = dst; + dest->dst_rtos = rtos; + dst_release(old_dst); +} + +static inline struct dst_entry * +__ip_vs_dst_check(struct ip_vs_dest *dest, u32 rtos, u32 cookie) +{ + struct dst_entry *dst = dest->dst_cache; + + if (!dst) + return NULL; + if ((dst->obsolete + || (dest->af == AF_INET && rtos != dest->dst_rtos)) && + dst->ops->check(dst, cookie) == NULL) { + dest->dst_cache = NULL; + dst_release(dst); + return NULL; + } + dst_hold(dst); + return dst; +} + +static struct rtable * +__ip_vs_get_out_rt(struct ip_vs_conn *cp, u32 rtos) +{ + struct rtable *rt; /* Route to the other host */ + struct ip_vs_dest *dest = cp->dest; + + if (dest) { + spin_lock(&dest->dst_lock); + if (!(rt = (struct rtable *) + __ip_vs_dst_check(dest, rtos, 0))) { + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip4_u = { + .daddr = dest->addr.ip, + .saddr = 0, + .tos = rtos, } }, + }; + + if (ip_route_output_key(&init_net, &rt, &fl)) { + spin_unlock(&dest->dst_lock); + IP_VS_DBG_RL("ip_route_output error, " + "dest: %u.%u.%u.%u\n", + NIPQUAD(dest->addr.ip)); + return NULL; + } + __ip_vs_dst_set(dest, rtos, dst_clone(&rt->u.dst)); + IP_VS_DBG(10, "new dst %u.%u.%u.%u, refcnt=%d, rtos=%X\n", + NIPQUAD(dest->addr.ip), + atomic_read(&rt->u.dst.__refcnt), rtos); + } + spin_unlock(&dest->dst_lock); + } else { + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip4_u = { + .daddr = cp->daddr.ip, + .saddr = 0, + .tos = rtos, } }, + }; + + if (ip_route_output_key(&init_net, &rt, &fl)) { + IP_VS_DBG_RL("ip_route_output error, dest: " + "%u.%u.%u.%u\n", NIPQUAD(cp->daddr.ip)); + return NULL; + } + } + + return rt; +} + +#ifdef CONFIG_IP_VS_IPV6 +static struct rt6_info * +__ip_vs_get_out_rt_v6(struct ip_vs_conn *cp) +{ + struct rt6_info *rt; /* Route to the other host */ + struct ip_vs_dest *dest = cp->dest; + + if (dest) { + spin_lock(&dest->dst_lock); + rt = (struct rt6_info *)__ip_vs_dst_check(dest, 0, 0); + if (!rt) { + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip6_u = { + .daddr = dest->addr.in6, + .saddr = { + .s6_addr32 = + { 0, 0, 0, 0 }, + }, + }, + }, + }; + + rt = (struct rt6_info *)ip6_route_output(&init_net, + NULL, &fl); + if (!rt) { + spin_unlock(&dest->dst_lock); + IP_VS_DBG_RL("ip6_route_output error, " + "dest: " NIP6_FMT "\n", + NIP6(dest->addr.in6)); + return NULL; + } + __ip_vs_dst_set(dest, 0, dst_clone(&rt->u.dst)); + IP_VS_DBG(10, "new dst " NIP6_FMT ", refcnt=%d\n", + NIP6(dest->addr.in6), + atomic_read(&rt->u.dst.__refcnt)); + } + spin_unlock(&dest->dst_lock); + } else { + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip6_u = { + .daddr = cp->daddr.in6, + .saddr = { + .s6_addr32 = { 0, 0, 0, 0 }, + }, + }, + }, + }; + + rt = (struct rt6_info *)ip6_route_output(&init_net, NULL, &fl); + if (!rt) { + IP_VS_DBG_RL("ip6_route_output error, dest: " + NIP6_FMT "\n", NIP6(cp->daddr.in6)); + return NULL; + } + } + + return rt; +} +#endif + + +/* + * Release dest->dst_cache before a dest is removed + */ +void +ip_vs_dst_reset(struct ip_vs_dest *dest) +{ + struct dst_entry *old_dst; + + old_dst = dest->dst_cache; + dest->dst_cache = NULL; + dst_release(old_dst); +} + +#define IP_VS_XMIT(pf, skb, rt) \ +do { \ + (skb)->ipvs_property = 1; \ + skb_forward_csum(skb); \ + NF_HOOK(pf, NF_INET_LOCAL_OUT, (skb), NULL, \ + (rt)->u.dst.dev, dst_output); \ +} while (0) + + +/* + * NULL transmitter (do nothing except return NF_ACCEPT) + */ +int +ip_vs_null_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + /* we do not touch skb and do not need pskb ptr */ + return NF_ACCEPT; +} + + +/* + * Bypass transmitter + * Let packets bypass the destination when the destination is not + * available, it may be only used in transparent cache cluster. + */ +int +ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rtable *rt; /* Route to the other host */ + struct iphdr *iph = ip_hdr(skb); + u8 tos = iph->tos; + int mtu; + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip4_u = { + .daddr = iph->daddr, + .saddr = 0, + .tos = RT_TOS(tos), } }, + }; + + EnterFunction(10); + + if (ip_route_output_key(&init_net, &rt, &fl)) { + IP_VS_DBG_RL("ip_vs_bypass_xmit(): ip_route_output error, " + "dest: %u.%u.%u.%u\n", NIPQUAD(iph->daddr)); + goto tx_error_icmp; + } + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF))) { + ip_rt_put(rt); + icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); + IP_VS_DBG_RL("ip_vs_bypass_xmit(): frag needed\n"); + goto tx_error; + } + + /* + * Call ip_send_check because we are not sure it is called + * after ip_defrag. Is copy-on-write needed? + */ + if (unlikely((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)) { + ip_rt_put(rt); + return NF_STOLEN; + } + ip_send_check(ip_hdr(skb)); + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET, skb, rt); + + LeaveFunction(10); + return NF_STOLEN; + + tx_error_icmp: + dst_link_failure(skb); + tx_error: + kfree_skb(skb); + LeaveFunction(10); + return NF_STOLEN; +} + +#ifdef CONFIG_IP_VS_IPV6 +int +ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rt6_info *rt; /* Route to the other host */ + struct ipv6hdr *iph = ipv6_hdr(skb); + int mtu; + struct flowi fl = { + .oif = 0, + .nl_u = { + .ip6_u = { + .daddr = iph->daddr, + .saddr = { .s6_addr32 = {0, 0, 0, 0} }, } }, + }; + + EnterFunction(10); + + rt = (struct rt6_info *)ip6_route_output(&init_net, NULL, &fl); + if (!rt) { + IP_VS_DBG_RL("ip_vs_bypass_xmit_v6(): ip6_route_output error, " + "dest: " NIP6_FMT "\n", NIP6(iph->daddr)); + goto tx_error_icmp; + } + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if (skb->len > mtu) { + dst_release(&rt->u.dst); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + IP_VS_DBG_RL("ip_vs_bypass_xmit_v6(): frag needed\n"); + goto tx_error; + } + + /* + * Call ip_send_check because we are not sure it is called + * after ip_defrag. Is copy-on-write needed? + */ + skb = skb_share_check(skb, GFP_ATOMIC); + if (unlikely(skb == NULL)) { + dst_release(&rt->u.dst); + return NF_STOLEN; + } + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET6, skb, rt); + + LeaveFunction(10); + return NF_STOLEN; + + tx_error_icmp: + dst_link_failure(skb); + tx_error: + kfree_skb(skb); + LeaveFunction(10); + return NF_STOLEN; +} +#endif + +/* + * NAT transmitter (only for outside-to-inside nat forwarding) + * Not used for related ICMP + */ +int +ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rtable *rt; /* Route to the other host */ + int mtu; + struct iphdr *iph = ip_hdr(skb); + + EnterFunction(10); + + /* check if it is a connection of no-client-port */ + if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT)) { + __be16 _pt, *p; + p = skb_header_pointer(skb, iph->ihl*4, sizeof(_pt), &_pt); + if (p == NULL) + goto tx_error; + ip_vs_conn_fill_cport(cp, *p); + IP_VS_DBG(10, "filled cport=%d\n", ntohs(*p)); + } + + if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(iph->tos)))) + goto tx_error_icmp; + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF))) { + ip_rt_put(rt); + icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); + IP_VS_DBG_RL_PKT(0, pp, skb, 0, "ip_vs_nat_xmit(): frag needed for"); + goto tx_error; + } + + /* copy-on-write the packet before mangling it */ + if (!skb_make_writable(skb, sizeof(struct iphdr))) + goto tx_error_put; + + if (skb_cow(skb, rt->u.dst.dev->hard_header_len)) + goto tx_error_put; + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* mangle the packet */ + if (pp->dnat_handler && !pp->dnat_handler(skb, pp, cp)) + goto tx_error; + ip_hdr(skb)->daddr = cp->daddr.ip; + ip_send_check(ip_hdr(skb)); + + IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT"); + + /* FIXME: when application helper enlarges the packet and the length + is larger than the MTU of outgoing device, there will be still + MTU problem. */ + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET, skb, rt); + + LeaveFunction(10); + return NF_STOLEN; + + tx_error_icmp: + dst_link_failure(skb); + tx_error: + LeaveFunction(10); + kfree_skb(skb); + return NF_STOLEN; + tx_error_put: + ip_rt_put(rt); + goto tx_error; +} + +#ifdef CONFIG_IP_VS_IPV6 +int +ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rt6_info *rt; /* Route to the other host */ + int mtu; + + EnterFunction(10); + + /* check if it is a connection of no-client-port */ + if (unlikely(cp->flags & IP_VS_CONN_F_NO_CPORT)) { + __be16 _pt, *p; + p = skb_header_pointer(skb, sizeof(struct ipv6hdr), + sizeof(_pt), &_pt); + if (p == NULL) + goto tx_error; + ip_vs_conn_fill_cport(cp, *p); + IP_VS_DBG(10, "filled cport=%d\n", ntohs(*p)); + } + + rt = __ip_vs_get_out_rt_v6(cp); + if (!rt) + goto tx_error_icmp; + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if (skb->len > mtu) { + dst_release(&rt->u.dst); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + IP_VS_DBG_RL_PKT(0, pp, skb, 0, + "ip_vs_nat_xmit_v6(): frag needed for"); + goto tx_error; + } + + /* copy-on-write the packet before mangling it */ + if (!skb_make_writable(skb, sizeof(struct ipv6hdr))) + goto tx_error_put; + + if (skb_cow(skb, rt->u.dst.dev->hard_header_len)) + goto tx_error_put; + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* mangle the packet */ + if (pp->dnat_handler && !pp->dnat_handler(skb, pp, cp)) + goto tx_error; + ipv6_hdr(skb)->daddr = cp->daddr.in6; + + IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT"); + + /* FIXME: when application helper enlarges the packet and the length + is larger than the MTU of outgoing device, there will be still + MTU problem. */ + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET6, skb, rt); + + LeaveFunction(10); + return NF_STOLEN; + +tx_error_icmp: + dst_link_failure(skb); +tx_error: + LeaveFunction(10); + kfree_skb(skb); + return NF_STOLEN; +tx_error_put: + dst_release(&rt->u.dst); + goto tx_error; +} +#endif + + +/* + * IP Tunneling transmitter + * + * This function encapsulates the packet in a new IP packet, its + * destination will be set to cp->daddr. Most code of this function + * is taken from ipip.c. + * + * It is used in VS/TUN cluster. The load balancer selects a real + * server from a cluster based on a scheduling algorithm, + * encapsulates the request packet and forwards it to the selected + * server. For example, all real servers are configured with + * "ifconfig tunl0 <Virtual IP Address> up". When the server receives + * the encapsulated packet, it will decapsulate the packet, processe + * the request and return the response packets directly to the client + * without passing the load balancer. This can greatly increase the + * scalability of virtual server. + * + * Used for ANY protocol + */ +int +ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rtable *rt; /* Route to the other host */ + struct net_device *tdev; /* Device to other host */ + struct iphdr *old_iph = ip_hdr(skb); + u8 tos = old_iph->tos; + __be16 df = old_iph->frag_off; + sk_buff_data_t old_transport_header = skb->transport_header; + struct iphdr *iph; /* Our new IP header */ + unsigned int max_headroom; /* The extra header space needed */ + int mtu; + + EnterFunction(10); + + if (skb->protocol != htons(ETH_P_IP)) { + IP_VS_DBG_RL("ip_vs_tunnel_xmit(): protocol error, " + "ETH_P_IP: %d, skb protocol: %d\n", + htons(ETH_P_IP), skb->protocol); + goto tx_error; + } + + if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(tos)))) + goto tx_error_icmp; + + tdev = rt->u.dst.dev; + + mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr); + if (mtu < 68) { + ip_rt_put(rt); + IP_VS_DBG_RL("ip_vs_tunnel_xmit(): mtu less than 68\n"); + goto tx_error; + } + if (skb->dst) + skb->dst->ops->update_pmtu(skb->dst, mtu); + + df |= (old_iph->frag_off & htons(IP_DF)); + + if ((old_iph->frag_off & htons(IP_DF)) + && mtu < ntohs(old_iph->tot_len)) { + icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); + ip_rt_put(rt); + IP_VS_DBG_RL("ip_vs_tunnel_xmit(): frag needed\n"); + goto tx_error; + } + + /* + * Okay, now see if we can stuff it in the buffer as-is. + */ + max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct iphdr); + + if (skb_headroom(skb) < max_headroom + || skb_cloned(skb) || skb_shared(skb)) { + struct sk_buff *new_skb = + skb_realloc_headroom(skb, max_headroom); + if (!new_skb) { + ip_rt_put(rt); + kfree_skb(skb); + IP_VS_ERR_RL("ip_vs_tunnel_xmit(): no memory\n"); + return NF_STOLEN; + } + kfree_skb(skb); + skb = new_skb; + old_iph = ip_hdr(skb); + } + + skb->transport_header = old_transport_header; + + /* fix old IP header checksum */ + ip_send_check(old_iph); + + skb_push(skb, sizeof(struct iphdr)); + skb_reset_network_header(skb); + memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* + * Push down and install the IPIP header. + */ + iph = ip_hdr(skb); + iph->version = 4; + iph->ihl = sizeof(struct iphdr)>>2; + iph->frag_off = df; + iph->protocol = IPPROTO_IPIP; + iph->tos = tos; + iph->daddr = rt->rt_dst; + iph->saddr = rt->rt_src; + iph->ttl = old_iph->ttl; + ip_select_ident(iph, &rt->u.dst, NULL); + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + ip_local_out(skb); + + LeaveFunction(10); + + return NF_STOLEN; + + tx_error_icmp: + dst_link_failure(skb); + tx_error: + kfree_skb(skb); + LeaveFunction(10); + return NF_STOLEN; +} + +#ifdef CONFIG_IP_VS_IPV6 +int +ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rt6_info *rt; /* Route to the other host */ + struct net_device *tdev; /* Device to other host */ + struct ipv6hdr *old_iph = ipv6_hdr(skb); + sk_buff_data_t old_transport_header = skb->transport_header; + struct ipv6hdr *iph; /* Our new IP header */ + unsigned int max_headroom; /* The extra header space needed */ + int mtu; + + EnterFunction(10); + + if (skb->protocol != htons(ETH_P_IPV6)) { + IP_VS_DBG_RL("ip_vs_tunnel_xmit_v6(): protocol error, " + "ETH_P_IPV6: %d, skb protocol: %d\n", + htons(ETH_P_IPV6), skb->protocol); + goto tx_error; + } + + rt = __ip_vs_get_out_rt_v6(cp); + if (!rt) + goto tx_error_icmp; + + tdev = rt->u.dst.dev; + + mtu = dst_mtu(&rt->u.dst) - sizeof(struct ipv6hdr); + /* TODO IPv6: do we need this check in IPv6? */ + if (mtu < 1280) { + dst_release(&rt->u.dst); + IP_VS_DBG_RL("ip_vs_tunnel_xmit_v6(): mtu less than 1280\n"); + goto tx_error; + } + if (skb->dst) + skb->dst->ops->update_pmtu(skb->dst, mtu); + + if (mtu < ntohs(old_iph->payload_len) + sizeof(struct ipv6hdr)) { + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + dst_release(&rt->u.dst); + IP_VS_DBG_RL("ip_vs_tunnel_xmit_v6(): frag needed\n"); + goto tx_error; + } + + /* + * Okay, now see if we can stuff it in the buffer as-is. + */ + max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct ipv6hdr); + + if (skb_headroom(skb) < max_headroom + || skb_cloned(skb) || skb_shared(skb)) { + struct sk_buff *new_skb = + skb_realloc_headroom(skb, max_headroom); + if (!new_skb) { + dst_release(&rt->u.dst); + kfree_skb(skb); + IP_VS_ERR_RL("ip_vs_tunnel_xmit_v6(): no memory\n"); + return NF_STOLEN; + } + kfree_skb(skb); + skb = new_skb; + old_iph = ipv6_hdr(skb); + } + + skb->transport_header = old_transport_header; + + skb_push(skb, sizeof(struct ipv6hdr)); + skb_reset_network_header(skb); + memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* + * Push down and install the IPIP header. + */ + iph = ipv6_hdr(skb); + iph->version = 6; + iph->nexthdr = IPPROTO_IPV6; + iph->payload_len = old_iph->payload_len + sizeof(old_iph); + iph->priority = old_iph->priority; + memset(&iph->flow_lbl, 0, sizeof(iph->flow_lbl)); + iph->daddr = rt->rt6i_dst.addr; + iph->saddr = cp->vaddr.in6; /* rt->rt6i_src.addr; */ + iph->hop_limit = old_iph->hop_limit; + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + ip6_local_out(skb); + + LeaveFunction(10); + + return NF_STOLEN; + +tx_error_icmp: + dst_link_failure(skb); +tx_error: + kfree_skb(skb); + LeaveFunction(10); + return NF_STOLEN; +} +#endif + + +/* + * Direct Routing transmitter + * Used for ANY protocol + */ +int +ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rtable *rt; /* Route to the other host */ + struct iphdr *iph = ip_hdr(skb); + int mtu; + + EnterFunction(10); + + if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(iph->tos)))) + goto tx_error_icmp; + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if ((iph->frag_off & htons(IP_DF)) && skb->len > mtu) { + icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); + ip_rt_put(rt); + IP_VS_DBG_RL("ip_vs_dr_xmit(): frag needed\n"); + goto tx_error; + } + + /* + * Call ip_send_check because we are not sure it is called + * after ip_defrag. Is copy-on-write needed? + */ + if (unlikely((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)) { + ip_rt_put(rt); + return NF_STOLEN; + } + ip_send_check(ip_hdr(skb)); + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET, skb, rt); + + LeaveFunction(10); + return NF_STOLEN; + + tx_error_icmp: + dst_link_failure(skb); + tx_error: + kfree_skb(skb); + LeaveFunction(10); + return NF_STOLEN; +} + +#ifdef CONFIG_IP_VS_IPV6 +int +ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) +{ + struct rt6_info *rt; /* Route to the other host */ + int mtu; + + EnterFunction(10); + + rt = __ip_vs_get_out_rt_v6(cp); + if (!rt) + goto tx_error_icmp; + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if (skb->len > mtu) { + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + dst_release(&rt->u.dst); + IP_VS_DBG_RL("ip_vs_dr_xmit_v6(): frag needed\n"); + goto tx_error; + } + + /* + * Call ip_send_check because we are not sure it is called + * after ip_defrag. Is copy-on-write needed? + */ + skb = skb_share_check(skb, GFP_ATOMIC); + if (unlikely(skb == NULL)) { + dst_release(&rt->u.dst); + return NF_STOLEN; + } + + /* drop old route */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET6, skb, rt); + + LeaveFunction(10); + return NF_STOLEN; + +tx_error_icmp: + dst_link_failure(skb); +tx_error: + kfree_skb(skb); + LeaveFunction(10); + return NF_STOLEN; +} +#endif + + +/* + * ICMP packet transmitter + * called by the ip_vs_in_icmp + */ +int +ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp, int offset) +{ + struct rtable *rt; /* Route to the other host */ + int mtu; + int rc; + + EnterFunction(10); + + /* The ICMP packet for VS/TUN, VS/DR and LOCALNODE will be + forwarded directly here, because there is no need to + translate address/port back */ + if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) { + if (cp->packet_xmit) + rc = cp->packet_xmit(skb, cp, pp); + else + rc = NF_ACCEPT; + /* do not touch skb anymore */ + atomic_inc(&cp->in_pkts); + goto out; + } + + /* + * mangle and send the packet here (only for VS/NAT) + */ + + if (!(rt = __ip_vs_get_out_rt(cp, RT_TOS(ip_hdr(skb)->tos)))) + goto tx_error_icmp; + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if ((skb->len > mtu) && (ip_hdr(skb)->frag_off & htons(IP_DF))) { + ip_rt_put(rt); + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); + IP_VS_DBG_RL("ip_vs_in_icmp(): frag needed\n"); + goto tx_error; + } + + /* copy-on-write the packet before mangling it */ + if (!skb_make_writable(skb, offset)) + goto tx_error_put; + + if (skb_cow(skb, rt->u.dst.dev->hard_header_len)) + goto tx_error_put; + + /* drop the old route when skb is not shared */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + ip_vs_nat_icmp(skb, pp, cp, 0); + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET, skb, rt); + + rc = NF_STOLEN; + goto out; + + tx_error_icmp: + dst_link_failure(skb); + tx_error: + dev_kfree_skb(skb); + rc = NF_STOLEN; + out: + LeaveFunction(10); + return rc; + tx_error_put: + ip_rt_put(rt); + goto tx_error; +} + +#ifdef CONFIG_IP_VS_IPV6 +int +ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, + struct ip_vs_protocol *pp, int offset) +{ + struct rt6_info *rt; /* Route to the other host */ + int mtu; + int rc; + + EnterFunction(10); + + /* The ICMP packet for VS/TUN, VS/DR and LOCALNODE will be + forwarded directly here, because there is no need to + translate address/port back */ + if (IP_VS_FWD_METHOD(cp) != IP_VS_CONN_F_MASQ) { + if (cp->packet_xmit) + rc = cp->packet_xmit(skb, cp, pp); + else + rc = NF_ACCEPT; + /* do not touch skb anymore */ + atomic_inc(&cp->in_pkts); + goto out; + } + + /* + * mangle and send the packet here (only for VS/NAT) + */ + + rt = __ip_vs_get_out_rt_v6(cp); + if (!rt) + goto tx_error_icmp; + + /* MTU checking */ + mtu = dst_mtu(&rt->u.dst); + if (skb->len > mtu) { + dst_release(&rt->u.dst); + icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, skb->dev); + IP_VS_DBG_RL("ip_vs_in_icmp(): frag needed\n"); + goto tx_error; + } + + /* copy-on-write the packet before mangling it */ + if (!skb_make_writable(skb, offset)) + goto tx_error_put; + + if (skb_cow(skb, rt->u.dst.dev->hard_header_len)) + goto tx_error_put; + + /* drop the old route when skb is not shared */ + dst_release(skb->dst); + skb->dst = &rt->u.dst; + + ip_vs_nat_icmp_v6(skb, pp, cp, 0); + + /* Another hack: avoid icmp_send in ip_fragment */ + skb->local_df = 1; + + IP_VS_XMIT(PF_INET6, skb, rt); + + rc = NF_STOLEN; + goto out; + +tx_error_icmp: + dst_link_failure(skb); +tx_error: + dev_kfree_skb(skb); + rc = NF_STOLEN; +out: + LeaveFunction(10); + return rc; +tx_error_put: + dst_release(&rt->u.dst); + goto tx_error; +} +#endif diff --git a/net/netfilter/nf_conntrack_acct.c b/net/netfilter/nf_conntrack_acct.c index 59bd8b903a1..03591d37b9c 100644 --- a/net/netfilter/nf_conntrack_acct.c +++ b/net/netfilter/nf_conntrack_acct.c @@ -22,19 +22,17 @@ #define NF_CT_ACCT_DEFAULT 0 #endif -int nf_ct_acct __read_mostly = NF_CT_ACCT_DEFAULT; -EXPORT_SYMBOL_GPL(nf_ct_acct); +static int nf_ct_acct __read_mostly = NF_CT_ACCT_DEFAULT; module_param_named(acct, nf_ct_acct, bool, 0644); MODULE_PARM_DESC(acct, "Enable connection tracking flow accounting."); #ifdef CONFIG_SYSCTL -static struct ctl_table_header *acct_sysctl_header; static struct ctl_table acct_sysctl_table[] = { { .ctl_name = CTL_UNNUMBERED, .procname = "nf_conntrack_acct", - .data = &nf_ct_acct, + .data = &init_net.ct.sysctl_acct, .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = &proc_dointvec, @@ -64,41 +62,87 @@ static struct nf_ct_ext_type acct_extend __read_mostly = { .id = NF_CT_EXT_ACCT, }; -int nf_conntrack_acct_init(void) +#ifdef CONFIG_SYSCTL +static int nf_conntrack_acct_init_sysctl(struct net *net) { - int ret; + struct ctl_table *table; -#ifdef CONFIG_NF_CT_ACCT - printk(KERN_WARNING "CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Plase use\n"); - printk(KERN_WARNING "nf_conntrack.acct=1 kernel paramater, acct=1 nf_conntrack module option or\n"); - printk(KERN_WARNING "sysctl net.netfilter.nf_conntrack_acct=1 to enable it.\n"); -#endif + table = kmemdup(acct_sysctl_table, sizeof(acct_sysctl_table), + GFP_KERNEL); + if (!table) + goto out; + + table[0].data = &net->ct.sysctl_acct; - ret = nf_ct_extend_register(&acct_extend); - if (ret < 0) { - printk(KERN_ERR "nf_conntrack_acct: Unable to register extension\n"); - return ret; + net->ct.acct_sysctl_header = register_net_sysctl_table(net, + nf_net_netfilter_sysctl_path, table); + if (!net->ct.acct_sysctl_header) { + printk(KERN_ERR "nf_conntrack_acct: can't register to sysctl.\n"); + goto out_register; } + return 0; -#ifdef CONFIG_SYSCTL - acct_sysctl_header = register_sysctl_paths(nf_net_netfilter_sysctl_path, - acct_sysctl_table); +out_register: + kfree(table); +out: + return -ENOMEM; +} - if (!acct_sysctl_header) { - nf_ct_extend_unregister(&acct_extend); +static void nf_conntrack_acct_fini_sysctl(struct net *net) +{ + struct ctl_table *table; - printk(KERN_ERR "nf_conntrack_acct: can't register to sysctl.\n"); - return -ENOMEM; - } + table = net->ct.acct_sysctl_header->ctl_table_arg; + unregister_net_sysctl_table(net->ct.acct_sysctl_header); + kfree(table); +} +#else +static int nf_conntrack_acct_init_sysctl(struct net *net) +{ + return 0; +} + +static void nf_conntrack_acct_fini_sysctl(struct net *net) +{ +} +#endif + +int nf_conntrack_acct_init(struct net *net) +{ + int ret; + + net->ct.sysctl_acct = nf_ct_acct; + + if (net_eq(net, &init_net)) { +#ifdef CONFIG_NF_CT_ACCT + printk(KERN_WARNING "CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Plase use\n"); + printk(KERN_WARNING "nf_conntrack.acct=1 kernel paramater, acct=1 nf_conntrack module option or\n"); + printk(KERN_WARNING "sysctl net.netfilter.nf_conntrack_acct=1 to enable it.\n"); #endif + ret = nf_ct_extend_register(&acct_extend); + if (ret < 0) { + printk(KERN_ERR "nf_conntrack_acct: Unable to register extension\n"); + goto out_extend_register; + } + } + + ret = nf_conntrack_acct_init_sysctl(net); + if (ret < 0) + goto out_sysctl; + return 0; + +out_sysctl: + if (net_eq(net, &init_net)) + nf_ct_extend_unregister(&acct_extend); +out_extend_register: + return ret; } -void nf_conntrack_acct_fini(void) +void nf_conntrack_acct_fini(struct net *net) { -#ifdef CONFIG_SYSCTL - unregister_sysctl_table(acct_sysctl_header); -#endif - nf_ct_extend_unregister(&acct_extend); + nf_conntrack_acct_fini_sysctl(net); + if (net_eq(net, &init_net)) + nf_ct_extend_unregister(&acct_extend); } diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 9d1830da8e8..27de3c7b006 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -44,30 +44,17 @@ DEFINE_SPINLOCK(nf_conntrack_lock); EXPORT_SYMBOL_GPL(nf_conntrack_lock); -/* nf_conntrack_standalone needs this */ -atomic_t nf_conntrack_count = ATOMIC_INIT(0); -EXPORT_SYMBOL_GPL(nf_conntrack_count); - unsigned int nf_conntrack_htable_size __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_htable_size); int nf_conntrack_max __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_max); -struct hlist_head *nf_conntrack_hash __read_mostly; -EXPORT_SYMBOL_GPL(nf_conntrack_hash); - struct nf_conn nf_conntrack_untracked __read_mostly; EXPORT_SYMBOL_GPL(nf_conntrack_untracked); -unsigned int nf_ct_log_invalid __read_mostly; -HLIST_HEAD(unconfirmed); -static int nf_conntrack_vmalloc __read_mostly; static struct kmem_cache *nf_conntrack_cachep __read_mostly; -DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); -EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat); - static int nf_conntrack_hash_rnd_initted; static unsigned int nf_conntrack_hash_rnd; @@ -180,6 +167,7 @@ static void destroy_conntrack(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; + struct net *net = nf_ct_net(ct); struct nf_conntrack_l4proto *l4proto; pr_debug("destroy_conntrack(%p)\n", ct); @@ -212,7 +200,7 @@ destroy_conntrack(struct nf_conntrack *nfct) hlist_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode); } - NF_CT_STAT_INC(delete); + NF_CT_STAT_INC(net, delete); spin_unlock_bh(&nf_conntrack_lock); if (ct->master) @@ -225,6 +213,7 @@ destroy_conntrack(struct nf_conntrack *nfct) static void death_by_timeout(unsigned long ul_conntrack) { struct nf_conn *ct = (void *)ul_conntrack; + struct net *net = nf_ct_net(ct); struct nf_conn_help *help = nfct_help(ct); struct nf_conntrack_helper *helper; @@ -239,14 +228,14 @@ static void death_by_timeout(unsigned long ul_conntrack) spin_lock_bh(&nf_conntrack_lock); /* Inside lock so preempt is disabled on module removal path. * Otherwise we can get spurious warnings. */ - NF_CT_STAT_INC(delete_list); + NF_CT_STAT_INC(net, delete_list); clean_from_lists(ct); spin_unlock_bh(&nf_conntrack_lock); nf_ct_put(ct); } struct nf_conntrack_tuple_hash * -__nf_conntrack_find(const struct nf_conntrack_tuple *tuple) +__nf_conntrack_find(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_tuple_hash *h; struct hlist_node *n; @@ -256,13 +245,13 @@ __nf_conntrack_find(const struct nf_conntrack_tuple *tuple) * at least once for the stats anyway. */ local_bh_disable(); - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], hnode) { + hlist_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnode) { if (nf_ct_tuple_equal(tuple, &h->tuple)) { - NF_CT_STAT_INC(found); + NF_CT_STAT_INC(net, found); local_bh_enable(); return h; } - NF_CT_STAT_INC(searched); + NF_CT_STAT_INC(net, searched); } local_bh_enable(); @@ -272,13 +261,13 @@ EXPORT_SYMBOL_GPL(__nf_conntrack_find); /* Find a connection corresponding to a tuple. */ struct nf_conntrack_tuple_hash * -nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple) +nf_conntrack_find_get(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; rcu_read_lock(); - h = __nf_conntrack_find(tuple); + h = __nf_conntrack_find(net, tuple); if (h) { ct = nf_ct_tuplehash_to_ctrack(h); if (unlikely(!atomic_inc_not_zero(&ct->ct_general.use))) @@ -294,10 +283,12 @@ static void __nf_conntrack_hash_insert(struct nf_conn *ct, unsigned int hash, unsigned int repl_hash) { + struct net *net = nf_ct_net(ct); + hlist_add_head_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode, - &nf_conntrack_hash[hash]); + &net->ct.hash[hash]); hlist_add_head_rcu(&ct->tuplehash[IP_CT_DIR_REPLY].hnode, - &nf_conntrack_hash[repl_hash]); + &net->ct.hash[repl_hash]); } void nf_conntrack_hash_insert(struct nf_conn *ct) @@ -323,8 +314,10 @@ __nf_conntrack_confirm(struct sk_buff *skb) struct nf_conn_help *help; struct hlist_node *n; enum ip_conntrack_info ctinfo; + struct net *net; ct = nf_ct_get(skb, &ctinfo); + net = nf_ct_net(ct); /* ipt_REJECT uses nf_conntrack_attach to attach related ICMP/TCP RST packets in other direction. Actual packet @@ -351,11 +344,11 @@ __nf_conntrack_confirm(struct sk_buff *skb) /* See if there's one in the list already, including reverse: NAT could have grabbed it without realizing, since we're not in the hash. If there is, we lost race. */ - hlist_for_each_entry(h, n, &nf_conntrack_hash[hash], hnode) + hlist_for_each_entry(h, n, &net->ct.hash[hash], hnode) if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, &h->tuple)) goto out; - hlist_for_each_entry(h, n, &nf_conntrack_hash[repl_hash], hnode) + hlist_for_each_entry(h, n, &net->ct.hash[repl_hash], hnode) if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple, &h->tuple)) goto out; @@ -371,22 +364,22 @@ __nf_conntrack_confirm(struct sk_buff *skb) add_timer(&ct->timeout); atomic_inc(&ct->ct_general.use); set_bit(IPS_CONFIRMED_BIT, &ct->status); - NF_CT_STAT_INC(insert); + NF_CT_STAT_INC(net, insert); spin_unlock_bh(&nf_conntrack_lock); help = nfct_help(ct); if (help && help->helper) - nf_conntrack_event_cache(IPCT_HELPER, skb); + nf_conntrack_event_cache(IPCT_HELPER, ct); #ifdef CONFIG_NF_NAT_NEEDED if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) || test_bit(IPS_DST_NAT_DONE_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_NATINFO, skb); + nf_conntrack_event_cache(IPCT_NATINFO, ct); #endif nf_conntrack_event_cache(master_ct(ct) ? - IPCT_RELATED : IPCT_NEW, skb); + IPCT_RELATED : IPCT_NEW, ct); return NF_ACCEPT; out: - NF_CT_STAT_INC(insert_failed); + NF_CT_STAT_INC(net, insert_failed); spin_unlock_bh(&nf_conntrack_lock); return NF_DROP; } @@ -398,6 +391,7 @@ int nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, const struct nf_conn *ignored_conntrack) { + struct net *net = nf_ct_net(ignored_conntrack); struct nf_conntrack_tuple_hash *h; struct hlist_node *n; unsigned int hash = hash_conntrack(tuple); @@ -406,14 +400,14 @@ nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, * least once for the stats anyway. */ rcu_read_lock_bh(); - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], hnode) { + hlist_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnode) { if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack && nf_ct_tuple_equal(tuple, &h->tuple)) { - NF_CT_STAT_INC(found); + NF_CT_STAT_INC(net, found); rcu_read_unlock_bh(); return 1; } - NF_CT_STAT_INC(searched); + NF_CT_STAT_INC(net, searched); } rcu_read_unlock_bh(); @@ -425,7 +419,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_tuple_taken); /* There's a small race here where we may free a just-assured connection. Too bad: we're in trouble anyway. */ -static noinline int early_drop(unsigned int hash) +static noinline int early_drop(struct net *net, unsigned int hash) { /* Use oldest entry, which is roughly LRU */ struct nf_conntrack_tuple_hash *h; @@ -436,7 +430,7 @@ static noinline int early_drop(unsigned int hash) rcu_read_lock(); for (i = 0; i < nf_conntrack_htable_size; i++) { - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], + hlist_for_each_entry_rcu(h, n, &net->ct.hash[hash], hnode) { tmp = nf_ct_tuplehash_to_ctrack(h); if (!test_bit(IPS_ASSURED_BIT, &tmp->status)) @@ -458,13 +452,14 @@ static noinline int early_drop(unsigned int hash) if (del_timer(&ct->timeout)) { death_by_timeout((unsigned long)ct); dropped = 1; - NF_CT_STAT_INC_ATOMIC(early_drop); + NF_CT_STAT_INC_ATOMIC(net, early_drop); } nf_ct_put(ct); return dropped; } -struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, +struct nf_conn *nf_conntrack_alloc(struct net *net, + const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, gfp_t gfp) { @@ -476,13 +471,13 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, } /* We don't want any race condition at early drop stage */ - atomic_inc(&nf_conntrack_count); + atomic_inc(&net->ct.count); if (nf_conntrack_max && - unlikely(atomic_read(&nf_conntrack_count) > nf_conntrack_max)) { + unlikely(atomic_read(&net->ct.count) > nf_conntrack_max)) { unsigned int hash = hash_conntrack(orig); - if (!early_drop(hash)) { - atomic_dec(&nf_conntrack_count); + if (!early_drop(net, hash)) { + atomic_dec(&net->ct.count); if (net_ratelimit()) printk(KERN_WARNING "nf_conntrack: table full, dropping" @@ -494,7 +489,7 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, ct = kmem_cache_zalloc(nf_conntrack_cachep, gfp); if (ct == NULL) { pr_debug("nf_conntrack_alloc: Can't alloc conntrack.\n"); - atomic_dec(&nf_conntrack_count); + atomic_dec(&net->ct.count); return ERR_PTR(-ENOMEM); } @@ -503,6 +498,9 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, ct->tuplehash[IP_CT_DIR_REPLY].tuple = *repl; /* Don't set timer yet: wait for confirmation */ setup_timer(&ct->timeout, death_by_timeout, (unsigned long)ct); +#ifdef CONFIG_NET_NS + ct->ct_net = net; +#endif INIT_RCU_HEAD(&ct->rcu); return ct; @@ -512,10 +510,11 @@ EXPORT_SYMBOL_GPL(nf_conntrack_alloc); static void nf_conntrack_free_rcu(struct rcu_head *head) { struct nf_conn *ct = container_of(head, struct nf_conn, rcu); + struct net *net = nf_ct_net(ct); nf_ct_ext_free(ct); kmem_cache_free(nf_conntrack_cachep, ct); - atomic_dec(&nf_conntrack_count); + atomic_dec(&net->ct.count); } void nf_conntrack_free(struct nf_conn *ct) @@ -528,7 +527,8 @@ EXPORT_SYMBOL_GPL(nf_conntrack_free); /* Allocate a new conntrack: we return -ENOMEM if classification failed due to stress. Otherwise it really is unclassifiable. */ static struct nf_conntrack_tuple_hash * -init_conntrack(const struct nf_conntrack_tuple *tuple, +init_conntrack(struct net *net, + const struct nf_conntrack_tuple *tuple, struct nf_conntrack_l3proto *l3proto, struct nf_conntrack_l4proto *l4proto, struct sk_buff *skb, @@ -544,7 +544,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, return NULL; } - ct = nf_conntrack_alloc(tuple, &repl_tuple, GFP_ATOMIC); + ct = nf_conntrack_alloc(net, tuple, &repl_tuple, GFP_ATOMIC); if (ct == NULL || IS_ERR(ct)) { pr_debug("Can't allocate conntrack.\n"); return (struct nf_conntrack_tuple_hash *)ct; @@ -559,7 +559,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, nf_ct_acct_ext_add(ct, GFP_ATOMIC); spin_lock_bh(&nf_conntrack_lock); - exp = nf_ct_find_expectation(tuple); + exp = nf_ct_find_expectation(net, tuple); if (exp) { pr_debug("conntrack: expectation arrives ct=%p exp=%p\n", ct, exp); @@ -579,7 +579,7 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, ct->secmark = exp->master->secmark; #endif nf_conntrack_get(&ct->master->ct_general); - NF_CT_STAT_INC(expect_new); + NF_CT_STAT_INC(net, expect_new); } else { struct nf_conntrack_helper *helper; @@ -589,11 +589,12 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, if (help) rcu_assign_pointer(help->helper, helper); } - NF_CT_STAT_INC(new); + NF_CT_STAT_INC(net, new); } /* Overload tuple linked list to put us in unconfirmed list. */ - hlist_add_head(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode, &unconfirmed); + hlist_add_head(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnode, + &net->ct.unconfirmed); spin_unlock_bh(&nf_conntrack_lock); @@ -608,7 +609,8 @@ init_conntrack(const struct nf_conntrack_tuple *tuple, /* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ static inline struct nf_conn * -resolve_normal_ct(struct sk_buff *skb, +resolve_normal_ct(struct net *net, + struct sk_buff *skb, unsigned int dataoff, u_int16_t l3num, u_int8_t protonum, @@ -629,9 +631,9 @@ resolve_normal_ct(struct sk_buff *skb, } /* look for tuple match */ - h = nf_conntrack_find_get(&tuple); + h = nf_conntrack_find_get(net, &tuple); if (!h) { - h = init_conntrack(&tuple, l3proto, l4proto, skb, dataoff); + h = init_conntrack(net, &tuple, l3proto, l4proto, skb, dataoff); if (!h) return NULL; if (IS_ERR(h)) @@ -665,7 +667,8 @@ resolve_normal_ct(struct sk_buff *skb, } unsigned int -nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb) +nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, + struct sk_buff *skb) { struct nf_conn *ct; enum ip_conntrack_info ctinfo; @@ -678,44 +681,46 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb) /* Previously seen (loopback or untracked)? Ignore. */ if (skb->nfct) { - NF_CT_STAT_INC_ATOMIC(ignore); + NF_CT_STAT_INC_ATOMIC(net, ignore); return NF_ACCEPT; } /* rcu_read_lock()ed by nf_hook_slow */ - l3proto = __nf_ct_l3proto_find((u_int16_t)pf); + l3proto = __nf_ct_l3proto_find(pf); ret = l3proto->get_l4proto(skb, skb_network_offset(skb), &dataoff, &protonum); if (ret <= 0) { pr_debug("not prepared to track yet or error occured\n"); - NF_CT_STAT_INC_ATOMIC(error); - NF_CT_STAT_INC_ATOMIC(invalid); + NF_CT_STAT_INC_ATOMIC(net, error); + NF_CT_STAT_INC_ATOMIC(net, invalid); return -ret; } - l4proto = __nf_ct_l4proto_find((u_int16_t)pf, protonum); + l4proto = __nf_ct_l4proto_find(pf, protonum); /* It may be an special packet, error, unclean... * inverse of the return code tells to the netfilter * core what to do with the packet. */ - if (l4proto->error != NULL && - (ret = l4proto->error(skb, dataoff, &ctinfo, pf, hooknum)) <= 0) { - NF_CT_STAT_INC_ATOMIC(error); - NF_CT_STAT_INC_ATOMIC(invalid); - return -ret; + if (l4proto->error != NULL) { + ret = l4proto->error(net, skb, dataoff, &ctinfo, pf, hooknum); + if (ret <= 0) { + NF_CT_STAT_INC_ATOMIC(net, error); + NF_CT_STAT_INC_ATOMIC(net, invalid); + return -ret; + } } - ct = resolve_normal_ct(skb, dataoff, pf, protonum, l3proto, l4proto, - &set_reply, &ctinfo); + ct = resolve_normal_ct(net, skb, dataoff, pf, protonum, + l3proto, l4proto, &set_reply, &ctinfo); if (!ct) { /* Not valid part of a connection */ - NF_CT_STAT_INC_ATOMIC(invalid); + NF_CT_STAT_INC_ATOMIC(net, invalid); return NF_ACCEPT; } if (IS_ERR(ct)) { /* Too stressed to deal. */ - NF_CT_STAT_INC_ATOMIC(drop); + NF_CT_STAT_INC_ATOMIC(net, drop); return NF_DROP; } @@ -728,12 +733,12 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb) pr_debug("nf_conntrack_in: Can't track with proto module\n"); nf_conntrack_put(skb->nfct); skb->nfct = NULL; - NF_CT_STAT_INC_ATOMIC(invalid); + NF_CT_STAT_INC_ATOMIC(net, invalid); return -ret; } if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_STATUS, ct); return ret; } @@ -846,7 +851,7 @@ acct: /* must be unlocked when calling event cache */ if (event) - nf_conntrack_event_cache(event, skb); + nf_conntrack_event_cache(event, ct); } EXPORT_SYMBOL_GPL(__nf_ct_refresh_acct); @@ -938,7 +943,7 @@ static void nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb) /* Bring out ya dead! */ static struct nf_conn * -get_next_corpse(int (*iter)(struct nf_conn *i, void *data), +get_next_corpse(struct net *net, int (*iter)(struct nf_conn *i, void *data), void *data, unsigned int *bucket) { struct nf_conntrack_tuple_hash *h; @@ -947,13 +952,13 @@ get_next_corpse(int (*iter)(struct nf_conn *i, void *data), spin_lock_bh(&nf_conntrack_lock); for (; *bucket < nf_conntrack_htable_size; (*bucket)++) { - hlist_for_each_entry(h, n, &nf_conntrack_hash[*bucket], hnode) { + hlist_for_each_entry(h, n, &net->ct.hash[*bucket], hnode) { ct = nf_ct_tuplehash_to_ctrack(h); if (iter(ct, data)) goto found; } } - hlist_for_each_entry(h, n, &unconfirmed, hnode) { + hlist_for_each_entry(h, n, &net->ct.unconfirmed, hnode) { ct = nf_ct_tuplehash_to_ctrack(h); if (iter(ct, data)) set_bit(IPS_DYING_BIT, &ct->status); @@ -966,13 +971,14 @@ found: return ct; } -void -nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) +void nf_ct_iterate_cleanup(struct net *net, + int (*iter)(struct nf_conn *i, void *data), + void *data) { struct nf_conn *ct; unsigned int bucket = 0; - while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) { + while ((ct = get_next_corpse(net, iter, data, &bucket)) != NULL) { /* Time to push up daises... */ if (del_timer(&ct->timeout)) death_by_timeout((unsigned long)ct); @@ -998,27 +1004,26 @@ void nf_ct_free_hashtable(struct hlist_head *hash, int vmalloced, unsigned int s } EXPORT_SYMBOL_GPL(nf_ct_free_hashtable); -void nf_conntrack_flush(void) +void nf_conntrack_flush(struct net *net) { - nf_ct_iterate_cleanup(kill_all, NULL); + nf_ct_iterate_cleanup(net, kill_all, NULL); } EXPORT_SYMBOL_GPL(nf_conntrack_flush); -/* Mishearing the voices in his head, our hero wonders how he's - supposed to kill the mall. */ -void nf_conntrack_cleanup(void) +static void nf_conntrack_cleanup_init_net(void) { - rcu_assign_pointer(ip_ct_attach, NULL); - - /* This makes sure all current packets have passed through - netfilter framework. Roll on, two-stage module - delete... */ - synchronize_net(); + nf_conntrack_helper_fini(); + nf_conntrack_proto_fini(); + kmem_cache_destroy(nf_conntrack_cachep); +} - nf_ct_event_cache_flush(); +static void nf_conntrack_cleanup_net(struct net *net) +{ + nf_ct_event_cache_flush(net); + nf_conntrack_ecache_fini(net); i_see_dead_people: - nf_conntrack_flush(); - if (atomic_read(&nf_conntrack_count) != 0) { + nf_conntrack_flush(net); + if (atomic_read(&net->ct.count) != 0) { schedule(); goto i_see_dead_people; } @@ -1026,16 +1031,31 @@ void nf_conntrack_cleanup(void) while (atomic_read(&nf_conntrack_untracked.ct_general.use) > 1) schedule(); - rcu_assign_pointer(nf_ct_destroy, NULL); - - kmem_cache_destroy(nf_conntrack_cachep); - nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_vmalloc, + nf_ct_free_hashtable(net->ct.hash, net->ct.hash_vmalloc, nf_conntrack_htable_size); + nf_conntrack_acct_fini(net); + nf_conntrack_expect_fini(net); + free_percpu(net->ct.stat); +} - nf_conntrack_acct_fini(); - nf_conntrack_expect_fini(); - nf_conntrack_helper_fini(); - nf_conntrack_proto_fini(); +/* Mishearing the voices in his head, our hero wonders how he's + supposed to kill the mall. */ +void nf_conntrack_cleanup(struct net *net) +{ + if (net_eq(net, &init_net)) + rcu_assign_pointer(ip_ct_attach, NULL); + + /* This makes sure all current packets have passed through + netfilter framework. Roll on, two-stage module + delete... */ + synchronize_net(); + + nf_conntrack_cleanup_net(net); + + if (net_eq(net, &init_net)) { + rcu_assign_pointer(nf_ct_destroy, NULL); + nf_conntrack_cleanup_init_net(); + } } struct hlist_head *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced) @@ -1094,8 +1114,8 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) */ spin_lock_bh(&nf_conntrack_lock); for (i = 0; i < nf_conntrack_htable_size; i++) { - while (!hlist_empty(&nf_conntrack_hash[i])) { - h = hlist_entry(nf_conntrack_hash[i].first, + while (!hlist_empty(&init_net.ct.hash[i])) { + h = hlist_entry(init_net.ct.hash[i].first, struct nf_conntrack_tuple_hash, hnode); hlist_del_rcu(&h->hnode); bucket = __hash_conntrack(&h->tuple, hashsize, rnd); @@ -1103,12 +1123,12 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) } } old_size = nf_conntrack_htable_size; - old_vmalloced = nf_conntrack_vmalloc; - old_hash = nf_conntrack_hash; + old_vmalloced = init_net.ct.hash_vmalloc; + old_hash = init_net.ct.hash; nf_conntrack_htable_size = hashsize; - nf_conntrack_vmalloc = vmalloced; - nf_conntrack_hash = hash; + init_net.ct.hash_vmalloc = vmalloced; + init_net.ct.hash = hash; nf_conntrack_hash_rnd = rnd; spin_unlock_bh(&nf_conntrack_lock); @@ -1120,7 +1140,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_set_hashsize); module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint, &nf_conntrack_htable_size, 0600); -int __init nf_conntrack_init(void) +static int nf_conntrack_init_init_net(void) { int max_factor = 8; int ret; @@ -1142,13 +1162,6 @@ int __init nf_conntrack_init(void) * entries. */ max_factor = 4; } - nf_conntrack_hash = nf_ct_alloc_hashtable(&nf_conntrack_htable_size, - &nf_conntrack_vmalloc); - if (!nf_conntrack_hash) { - printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); - goto err_out; - } - nf_conntrack_max = max_factor * nf_conntrack_htable_size; printk("nf_conntrack version %s (%u buckets, %d max)\n", @@ -1160,48 +1173,103 @@ int __init nf_conntrack_init(void) 0, 0, NULL); if (!nf_conntrack_cachep) { printk(KERN_ERR "Unable to create nf_conn slab cache\n"); - goto err_free_hash; + ret = -ENOMEM; + goto err_cache; } ret = nf_conntrack_proto_init(); if (ret < 0) - goto err_free_conntrack_slab; - - ret = nf_conntrack_expect_init(); - if (ret < 0) - goto out_fini_proto; + goto err_proto; ret = nf_conntrack_helper_init(); if (ret < 0) - goto out_fini_expect; + goto err_helper; - ret = nf_conntrack_acct_init(); - if (ret < 0) - goto out_fini_helper; + return 0; - /* For use by REJECT target */ - rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach); - rcu_assign_pointer(nf_ct_destroy, destroy_conntrack); +err_helper: + nf_conntrack_proto_fini(); +err_proto: + kmem_cache_destroy(nf_conntrack_cachep); +err_cache: + return ret; +} + +static int nf_conntrack_init_net(struct net *net) +{ + int ret; + + atomic_set(&net->ct.count, 0); + INIT_HLIST_HEAD(&net->ct.unconfirmed); + net->ct.stat = alloc_percpu(struct ip_conntrack_stat); + if (!net->ct.stat) { + ret = -ENOMEM; + goto err_stat; + } + ret = nf_conntrack_ecache_init(net); + if (ret < 0) + goto err_ecache; + net->ct.hash = nf_ct_alloc_hashtable(&nf_conntrack_htable_size, + &net->ct.hash_vmalloc); + if (!net->ct.hash) { + ret = -ENOMEM; + printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); + goto err_hash; + } + ret = nf_conntrack_expect_init(net); + if (ret < 0) + goto err_expect; + ret = nf_conntrack_acct_init(net); + if (ret < 0) + goto err_acct; /* Set up fake conntrack: - to never be deleted, not in any hashes */ +#ifdef CONFIG_NET_NS + nf_conntrack_untracked.ct_net = &init_net; +#endif atomic_set(&nf_conntrack_untracked.ct_general.use, 1); /* - and look it like as a confirmed connection */ set_bit(IPS_CONFIRMED_BIT, &nf_conntrack_untracked.status); - return ret; + return 0; -out_fini_helper: - nf_conntrack_helper_fini(); -out_fini_expect: - nf_conntrack_expect_fini(); -out_fini_proto: - nf_conntrack_proto_fini(); -err_free_conntrack_slab: - kmem_cache_destroy(nf_conntrack_cachep); -err_free_hash: - nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_vmalloc, +err_acct: + nf_conntrack_expect_fini(net); +err_expect: + nf_ct_free_hashtable(net->ct.hash, net->ct.hash_vmalloc, nf_conntrack_htable_size); -err_out: - return -ENOMEM; +err_hash: + nf_conntrack_ecache_fini(net); +err_ecache: + free_percpu(net->ct.stat); +err_stat: + return ret; +} + +int nf_conntrack_init(struct net *net) +{ + int ret; + + if (net_eq(net, &init_net)) { + ret = nf_conntrack_init_init_net(); + if (ret < 0) + goto out_init_net; + } + ret = nf_conntrack_init_net(net); + if (ret < 0) + goto out_net; + + if (net_eq(net, &init_net)) { + /* For use by REJECT target */ + rcu_assign_pointer(ip_ct_attach, nf_conntrack_attach); + rcu_assign_pointer(nf_ct_destroy, destroy_conntrack); + } + return 0; + +out_net: + if (net_eq(net, &init_net)) + nf_conntrack_cleanup_init_net(); +out_init_net: + return ret; } diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c index 83c41ac3505..a5f5e2e65d1 100644 --- a/net/netfilter/nf_conntrack_ecache.c +++ b/net/netfilter/nf_conntrack_ecache.c @@ -29,9 +29,6 @@ EXPORT_SYMBOL_GPL(nf_conntrack_chain); ATOMIC_NOTIFIER_HEAD(nf_ct_expect_chain); EXPORT_SYMBOL_GPL(nf_ct_expect_chain); -DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); -EXPORT_PER_CPU_SYMBOL_GPL(nf_conntrack_ecache); - /* deliver cached events and clear cache entry - must be called with locally * disabled softirqs */ static inline void @@ -51,10 +48,11 @@ __nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) * by code prior to async packet handling for freeing the skb */ void nf_ct_deliver_cached_events(const struct nf_conn *ct) { + struct net *net = nf_ct_net(ct); struct nf_conntrack_ecache *ecache; local_bh_disable(); - ecache = &__get_cpu_var(nf_conntrack_ecache); + ecache = per_cpu_ptr(net->ct.ecache, raw_smp_processor_id()); if (ecache->ct == ct) __nf_ct_deliver_cached_events(ecache); local_bh_enable(); @@ -64,10 +62,11 @@ EXPORT_SYMBOL_GPL(nf_ct_deliver_cached_events); /* Deliver cached events for old pending events, if current conntrack != old */ void __nf_ct_event_cache_init(struct nf_conn *ct) { + struct net *net = nf_ct_net(ct); struct nf_conntrack_ecache *ecache; /* take care of delivering potentially old events */ - ecache = &__get_cpu_var(nf_conntrack_ecache); + ecache = per_cpu_ptr(net->ct.ecache, raw_smp_processor_id()); BUG_ON(ecache->ct == ct); if (ecache->ct) __nf_ct_deliver_cached_events(ecache); @@ -79,18 +78,31 @@ EXPORT_SYMBOL_GPL(__nf_ct_event_cache_init); /* flush the event cache - touches other CPU's data and must not be called * while packets are still passing through the code */ -void nf_ct_event_cache_flush(void) +void nf_ct_event_cache_flush(struct net *net) { struct nf_conntrack_ecache *ecache; int cpu; for_each_possible_cpu(cpu) { - ecache = &per_cpu(nf_conntrack_ecache, cpu); + ecache = per_cpu_ptr(net->ct.ecache, cpu); if (ecache->ct) nf_ct_put(ecache->ct); } } +int nf_conntrack_ecache_init(struct net *net) +{ + net->ct.ecache = alloc_percpu(struct nf_conntrack_ecache); + if (!net->ct.ecache) + return -ENOMEM; + return 0; +} + +void nf_conntrack_ecache_fini(struct net *net) +{ + free_percpu(net->ct.ecache); +} + int nf_conntrack_register_notifier(struct notifier_block *nb) { return atomic_notifier_chain_register(&nf_conntrack_chain, nb); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index e8f0dead267..37a703bc3b8 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -28,17 +28,12 @@ #include <net/netfilter/nf_conntrack_helper.h> #include <net/netfilter/nf_conntrack_tuple.h> -struct hlist_head *nf_ct_expect_hash __read_mostly; -EXPORT_SYMBOL_GPL(nf_ct_expect_hash); - unsigned int nf_ct_expect_hsize __read_mostly; EXPORT_SYMBOL_GPL(nf_ct_expect_hsize); static unsigned int nf_ct_expect_hash_rnd __read_mostly; -static unsigned int nf_ct_expect_count; unsigned int nf_ct_expect_max __read_mostly; static int nf_ct_expect_hash_rnd_initted __read_mostly; -static int nf_ct_expect_vmalloc; static struct kmem_cache *nf_ct_expect_cachep __read_mostly; @@ -46,18 +41,19 @@ static struct kmem_cache *nf_ct_expect_cachep __read_mostly; void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) { struct nf_conn_help *master_help = nfct_help(exp->master); + struct net *net = nf_ct_exp_net(exp); NF_CT_ASSERT(master_help); NF_CT_ASSERT(!timer_pending(&exp->timeout)); hlist_del_rcu(&exp->hnode); - nf_ct_expect_count--; + net->ct.expect_count--; hlist_del(&exp->lnode); master_help->expecting[exp->class]--; nf_ct_expect_put(exp); - NF_CT_STAT_INC(expect_delete); + NF_CT_STAT_INC(net, expect_delete); } EXPORT_SYMBOL_GPL(nf_ct_unlink_expect); @@ -87,17 +83,17 @@ static unsigned int nf_ct_expect_dst_hash(const struct nf_conntrack_tuple *tuple } struct nf_conntrack_expect * -__nf_ct_expect_find(const struct nf_conntrack_tuple *tuple) +__nf_ct_expect_find(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i; struct hlist_node *n; unsigned int h; - if (!nf_ct_expect_count) + if (!net->ct.expect_count) return NULL; h = nf_ct_expect_dst_hash(tuple); - hlist_for_each_entry_rcu(i, n, &nf_ct_expect_hash[h], hnode) { + hlist_for_each_entry_rcu(i, n, &net->ct.expect_hash[h], hnode) { if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) return i; } @@ -107,12 +103,12 @@ EXPORT_SYMBOL_GPL(__nf_ct_expect_find); /* Just find a expectation corresponding to a tuple. */ struct nf_conntrack_expect * -nf_ct_expect_find_get(const struct nf_conntrack_tuple *tuple) +nf_ct_expect_find_get(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i; rcu_read_lock(); - i = __nf_ct_expect_find(tuple); + i = __nf_ct_expect_find(net, tuple); if (i && !atomic_inc_not_zero(&i->use)) i = NULL; rcu_read_unlock(); @@ -124,17 +120,17 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_find_get); /* If an expectation for this connection is found, it gets delete from * global list then returned. */ struct nf_conntrack_expect * -nf_ct_find_expectation(const struct nf_conntrack_tuple *tuple) +nf_ct_find_expectation(struct net *net, const struct nf_conntrack_tuple *tuple) { struct nf_conntrack_expect *i, *exp = NULL; struct hlist_node *n; unsigned int h; - if (!nf_ct_expect_count) + if (!net->ct.expect_count) return NULL; h = nf_ct_expect_dst_hash(tuple); - hlist_for_each_entry(i, n, &nf_ct_expect_hash[h], hnode) { + hlist_for_each_entry(i, n, &net->ct.expect_hash[h], hnode) { if (!(i->flags & NF_CT_EXPECT_INACTIVE) && nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) { exp = i; @@ -241,7 +237,7 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me) EXPORT_SYMBOL_GPL(nf_ct_expect_alloc); void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, - int family, + u_int8_t family, const union nf_inet_addr *saddr, const union nf_inet_addr *daddr, u_int8_t proto, const __be16 *src, const __be16 *dst) @@ -311,6 +307,7 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_put); static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) { struct nf_conn_help *master_help = nfct_help(exp->master); + struct net *net = nf_ct_exp_net(exp); const struct nf_conntrack_expect_policy *p; unsigned int h = nf_ct_expect_dst_hash(&exp->tuple); @@ -319,8 +316,8 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) hlist_add_head(&exp->lnode, &master_help->expectations); master_help->expecting[exp->class]++; - hlist_add_head_rcu(&exp->hnode, &nf_ct_expect_hash[h]); - nf_ct_expect_count++; + hlist_add_head_rcu(&exp->hnode, &net->ct.expect_hash[h]); + net->ct.expect_count++; setup_timer(&exp->timeout, nf_ct_expectation_timed_out, (unsigned long)exp); @@ -329,7 +326,7 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) add_timer(&exp->timeout); atomic_inc(&exp->use); - NF_CT_STAT_INC(expect_create); + NF_CT_STAT_INC(net, expect_create); } /* Race with expectations being used means we could have none to find; OK. */ @@ -371,6 +368,7 @@ int nf_ct_expect_related(struct nf_conntrack_expect *expect) struct nf_conntrack_expect *i; struct nf_conn *master = expect->master; struct nf_conn_help *master_help = nfct_help(master); + struct net *net = nf_ct_exp_net(expect); struct hlist_node *n; unsigned int h; int ret; @@ -383,7 +381,7 @@ int nf_ct_expect_related(struct nf_conntrack_expect *expect) goto out; } h = nf_ct_expect_dst_hash(&expect->tuple); - hlist_for_each_entry(i, n, &nf_ct_expect_hash[h], hnode) { + hlist_for_each_entry(i, n, &net->ct.expect_hash[h], hnode) { if (expect_matches(i, expect)) { /* Refresh timer: if it's dying, ignore.. */ if (refresh_timer(i)) { @@ -406,7 +404,7 @@ int nf_ct_expect_related(struct nf_conntrack_expect *expect) } } - if (nf_ct_expect_count >= nf_ct_expect_max) { + if (net->ct.expect_count >= nf_ct_expect_max) { if (net_ratelimit()) printk(KERN_WARNING "nf_conntrack: expectation table full\n"); @@ -425,16 +423,18 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_related); #ifdef CONFIG_PROC_FS struct ct_expect_iter_state { + struct seq_net_private p; unsigned int bucket; }; static struct hlist_node *ct_expect_get_first(struct seq_file *seq) { + struct net *net = seq_file_net(seq); struct ct_expect_iter_state *st = seq->private; struct hlist_node *n; for (st->bucket = 0; st->bucket < nf_ct_expect_hsize; st->bucket++) { - n = rcu_dereference(nf_ct_expect_hash[st->bucket].first); + n = rcu_dereference(net->ct.expect_hash[st->bucket].first); if (n) return n; } @@ -444,13 +444,14 @@ static struct hlist_node *ct_expect_get_first(struct seq_file *seq) static struct hlist_node *ct_expect_get_next(struct seq_file *seq, struct hlist_node *head) { + struct net *net = seq_file_net(seq); struct ct_expect_iter_state *st = seq->private; head = rcu_dereference(head->next); while (head == NULL) { if (++st->bucket >= nf_ct_expect_hsize) return NULL; - head = rcu_dereference(nf_ct_expect_hash[st->bucket].first); + head = rcu_dereference(net->ct.expect_hash[st->bucket].first); } return head; } @@ -524,7 +525,7 @@ static const struct seq_operations exp_seq_ops = { static int exp_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &exp_seq_ops, + return seq_open_net(inode, file, &exp_seq_ops, sizeof(struct ct_expect_iter_state)); } @@ -533,72 +534,79 @@ static const struct file_operations exp_file_ops = { .open = exp_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; #endif /* CONFIG_PROC_FS */ -static int __init exp_proc_init(void) +static int exp_proc_init(struct net *net) { #ifdef CONFIG_PROC_FS struct proc_dir_entry *proc; - proc = proc_net_fops_create(&init_net, "nf_conntrack_expect", 0440, &exp_file_ops); + proc = proc_net_fops_create(net, "nf_conntrack_expect", 0440, &exp_file_ops); if (!proc) return -ENOMEM; #endif /* CONFIG_PROC_FS */ return 0; } -static void exp_proc_remove(void) +static void exp_proc_remove(struct net *net) { #ifdef CONFIG_PROC_FS - proc_net_remove(&init_net, "nf_conntrack_expect"); + proc_net_remove(net, "nf_conntrack_expect"); #endif /* CONFIG_PROC_FS */ } module_param_named(expect_hashsize, nf_ct_expect_hsize, uint, 0600); -int __init nf_conntrack_expect_init(void) +int nf_conntrack_expect_init(struct net *net) { int err = -ENOMEM; - if (!nf_ct_expect_hsize) { - nf_ct_expect_hsize = nf_conntrack_htable_size / 256; - if (!nf_ct_expect_hsize) - nf_ct_expect_hsize = 1; + if (net_eq(net, &init_net)) { + if (!nf_ct_expect_hsize) { + nf_ct_expect_hsize = nf_conntrack_htable_size / 256; + if (!nf_ct_expect_hsize) + nf_ct_expect_hsize = 1; + } + nf_ct_expect_max = nf_ct_expect_hsize * 4; } - nf_ct_expect_max = nf_ct_expect_hsize * 4; - nf_ct_expect_hash = nf_ct_alloc_hashtable(&nf_ct_expect_hsize, - &nf_ct_expect_vmalloc); - if (nf_ct_expect_hash == NULL) + net->ct.expect_count = 0; + net->ct.expect_hash = nf_ct_alloc_hashtable(&nf_ct_expect_hsize, + &net->ct.expect_vmalloc); + if (net->ct.expect_hash == NULL) goto err1; - nf_ct_expect_cachep = kmem_cache_create("nf_conntrack_expect", + if (net_eq(net, &init_net)) { + nf_ct_expect_cachep = kmem_cache_create("nf_conntrack_expect", sizeof(struct nf_conntrack_expect), 0, 0, NULL); - if (!nf_ct_expect_cachep) - goto err2; + if (!nf_ct_expect_cachep) + goto err2; + } - err = exp_proc_init(); + err = exp_proc_init(net); if (err < 0) goto err3; return 0; err3: - kmem_cache_destroy(nf_ct_expect_cachep); + if (net_eq(net, &init_net)) + kmem_cache_destroy(nf_ct_expect_cachep); err2: - nf_ct_free_hashtable(nf_ct_expect_hash, nf_ct_expect_vmalloc, + nf_ct_free_hashtable(net->ct.expect_hash, net->ct.expect_vmalloc, nf_ct_expect_hsize); err1: return err; } -void nf_conntrack_expect_fini(void) +void nf_conntrack_expect_fini(struct net *net) { - exp_proc_remove(); - kmem_cache_destroy(nf_ct_expect_cachep); - nf_ct_free_hashtable(nf_ct_expect_hash, nf_ct_expect_vmalloc, + exp_proc_remove(net); + if (net_eq(net, &init_net)) + kmem_cache_destroy(nf_ct_expect_cachep); + nf_ct_free_hashtable(net->ct.expect_hash, net->ct.expect_vmalloc, nf_ct_expect_hsize); } diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index bb20672fe03..4f7107107e9 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -318,7 +318,8 @@ static int find_nl_seq(u32 seq, const struct nf_ct_ftp_master *info, int dir) } /* We don't update if it's older than what we have. */ -static void update_nl_seq(u32 nl_seq, struct nf_ct_ftp_master *info, int dir, +static void update_nl_seq(struct nf_conn *ct, u32 nl_seq, + struct nf_ct_ftp_master *info, int dir, struct sk_buff *skb) { unsigned int i, oldest = NUM_SEQ_TO_REMEMBER; @@ -336,11 +337,11 @@ static void update_nl_seq(u32 nl_seq, struct nf_ct_ftp_master *info, int dir, if (info->seq_aft_nl_num[dir] < NUM_SEQ_TO_REMEMBER) { info->seq_aft_nl[dir][info->seq_aft_nl_num[dir]++] = nl_seq; - nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, skb); + nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, ct); } else if (oldest != NUM_SEQ_TO_REMEMBER && after(nl_seq, info->seq_aft_nl[dir][oldest])) { info->seq_aft_nl[dir][oldest] = nl_seq; - nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, skb); + nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, ct); } } @@ -509,7 +510,7 @@ out_update_nl: /* Now if this ends in \n, update ftp info. Seq may have been * adjusted by NAT code. */ if (ends_in_nl) - update_nl_seq(seq, ct_ftp_info, dir, skb); + update_nl_seq(ct, seq, ct_ftp_info, dir, skb); out: spin_unlock_bh(&nf_ftp_lock); return ret; diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index 2f83c158934..c1504f71cdf 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -709,7 +709,8 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct, /* If the calling party is on the same side of the forward-to party, * we don't need to track the second call */ static int callforward_do_filter(const union nf_inet_addr *src, - const union nf_inet_addr *dst, int family) + const union nf_inet_addr *dst, + u_int8_t family) { const struct nf_afinfo *afinfo; struct flowi fl1, fl2; @@ -1209,6 +1210,7 @@ static struct nf_conntrack_expect *find_expect(struct nf_conn *ct, union nf_inet_addr *addr, __be16 port) { + struct net *net = nf_ct_net(ct); struct nf_conntrack_expect *exp; struct nf_conntrack_tuple tuple; @@ -1218,7 +1220,7 @@ static struct nf_conntrack_expect *find_expect(struct nf_conn *ct, tuple.dst.u.tcp.port = port; tuple.dst.protonum = IPPROTO_TCP; - exp = __nf_ct_expect_find(&tuple); + exp = __nf_ct_expect_find(net, &tuple); if (exp && exp->master == ct) return exp; return NULL; diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 8e0b4c8f62a..9c06b9f86ad 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -123,29 +123,18 @@ int nf_conntrack_helper_register(struct nf_conntrack_helper *me) } EXPORT_SYMBOL_GPL(nf_conntrack_helper_register); -void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +static void __nf_conntrack_helper_unregister(struct nf_conntrack_helper *me, + struct net *net) { struct nf_conntrack_tuple_hash *h; struct nf_conntrack_expect *exp; const struct hlist_node *n, *next; unsigned int i; - mutex_lock(&nf_ct_helper_mutex); - hlist_del_rcu(&me->hnode); - nf_ct_helper_count--; - mutex_unlock(&nf_ct_helper_mutex); - - /* Make sure every nothing is still using the helper unless its a - * connection in the hash. - */ - synchronize_rcu(); - - spin_lock_bh(&nf_conntrack_lock); - /* Get rid of expectations */ for (i = 0; i < nf_ct_expect_hsize; i++) { hlist_for_each_entry_safe(exp, n, next, - &nf_ct_expect_hash[i], hnode) { + &net->ct.expect_hash[i], hnode) { struct nf_conn_help *help = nfct_help(exp->master); if ((help->helper == me || exp->helper == me) && del_timer(&exp->timeout)) { @@ -156,12 +145,31 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) } /* Get rid of expecteds, set helpers to NULL. */ - hlist_for_each_entry(h, n, &unconfirmed, hnode) + hlist_for_each_entry(h, n, &net->ct.unconfirmed, hnode) unhelp(h, me); for (i = 0; i < nf_conntrack_htable_size; i++) { - hlist_for_each_entry(h, n, &nf_conntrack_hash[i], hnode) + hlist_for_each_entry(h, n, &net->ct.hash[i], hnode) unhelp(h, me); } +} + +void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +{ + struct net *net; + + mutex_lock(&nf_ct_helper_mutex); + hlist_del_rcu(&me->hnode); + nf_ct_helper_count--; + mutex_unlock(&nf_ct_helper_mutex); + + /* Make sure every nothing is still using the helper unless its a + * connection in the hash. + */ + synchronize_rcu(); + + spin_lock_bh(&nf_conntrack_lock); + for_each_net(net) + __nf_conntrack_helper_unregister(me, net); spin_unlock_bh(&nf_conntrack_lock); } EXPORT_SYMBOL_GPL(nf_conntrack_helper_unregister); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index a8752031adc..cadfd15b44f 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -549,7 +549,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) last = (struct nf_conn *)cb->args[1]; for (; cb->args[0] < nf_conntrack_htable_size; cb->args[0]++) { restart: - hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[cb->args[0]], + hlist_for_each_entry_rcu(h, n, &init_net.ct.hash[cb->args[0]], hnode) { if (NF_CT_DIRECTION(h) != IP_CT_DIR_ORIGINAL) continue; @@ -794,14 +794,14 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, err = ctnetlink_parse_tuple(cda, &tuple, CTA_TUPLE_REPLY, u3); else { /* Flush the whole table */ - nf_conntrack_flush(); + nf_conntrack_flush(&init_net); return 0; } if (err < 0) return err; - h = nf_conntrack_find_get(&tuple); + h = nf_conntrack_find_get(&init_net, &tuple); if (!h) return -ENOENT; @@ -847,7 +847,7 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - h = nf_conntrack_find_get(&tuple); + h = nf_conntrack_find_get(&init_net, &tuple); if (!h) return -ENOENT; @@ -1125,7 +1125,7 @@ ctnetlink_create_conntrack(struct nlattr *cda[], struct nf_conn_help *help; struct nf_conntrack_helper *helper; - ct = nf_conntrack_alloc(otuple, rtuple, GFP_KERNEL); + ct = nf_conntrack_alloc(&init_net, otuple, rtuple, GFP_KERNEL); if (ct == NULL || IS_ERR(ct)) return -ENOMEM; @@ -1213,9 +1213,9 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, spin_lock_bh(&nf_conntrack_lock); if (cda[CTA_TUPLE_ORIG]) - h = __nf_conntrack_find(&otuple); + h = __nf_conntrack_find(&init_net, &otuple); else if (cda[CTA_TUPLE_REPLY]) - h = __nf_conntrack_find(&rtuple); + h = __nf_conntrack_find(&init_net, &rtuple); if (h == NULL) { struct nf_conntrack_tuple master; @@ -1230,7 +1230,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, if (err < 0) goto out_unlock; - master_h = __nf_conntrack_find(&master); + master_h = __nf_conntrack_find(&init_net, &master); if (master_h == NULL) { err = -ENOENT; goto out_unlock; @@ -1458,6 +1458,7 @@ static int ctnetlink_exp_done(struct netlink_callback *cb) static int ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) { + struct net *net = &init_net; struct nf_conntrack_expect *exp, *last; struct nfgenmsg *nfmsg = NLMSG_DATA(cb->nlh); struct hlist_node *n; @@ -1467,7 +1468,7 @@ ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) last = (struct nf_conntrack_expect *)cb->args[1]; for (; cb->args[0] < nf_ct_expect_hsize; cb->args[0]++) { restart: - hlist_for_each_entry(exp, n, &nf_ct_expect_hash[cb->args[0]], + hlist_for_each_entry(exp, n, &net->ct.expect_hash[cb->args[0]], hnode) { if (l3proto && exp->tuple.src.l3num != l3proto) continue; @@ -1529,7 +1530,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - exp = nf_ct_expect_find_get(&tuple); + exp = nf_ct_expect_find_get(&init_net, &tuple); if (!exp) return -ENOENT; @@ -1583,7 +1584,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return err; /* bump usage count to 2 */ - exp = nf_ct_expect_find_get(&tuple); + exp = nf_ct_expect_find_get(&init_net, &tuple); if (!exp) return -ENOENT; @@ -1613,7 +1614,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, } for (i = 0; i < nf_ct_expect_hsize; i++) { hlist_for_each_entry_safe(exp, n, next, - &nf_ct_expect_hash[i], + &init_net.ct.expect_hash[i], hnode) { m_help = nfct_help(exp->master); if (m_help->helper == h @@ -1629,7 +1630,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, spin_lock_bh(&nf_conntrack_lock); for (i = 0; i < nf_ct_expect_hsize; i++) { hlist_for_each_entry_safe(exp, n, next, - &nf_ct_expect_hash[i], + &init_net.ct.expect_hash[i], hnode) { if (del_timer(&exp->timeout)) { nf_ct_unlink_expect(exp); @@ -1670,7 +1671,7 @@ ctnetlink_create_expect(struct nlattr *cda[], u_int8_t u3) return err; /* Look for master conntrack of this expectation */ - h = nf_conntrack_find_get(&master_tuple); + h = nf_conntrack_find_get(&init_net, &master_tuple); if (!h) return -ENOENT; ct = nf_ct_tuplehash_to_ctrack(h); @@ -1724,7 +1725,7 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, return err; spin_lock_bh(&nf_conntrack_lock); - exp = __nf_ct_expect_find(&tuple); + exp = __nf_ct_expect_find(&init_net, &tuple); if (!exp) { spin_unlock_bh(&nf_conntrack_lock); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 97e54b0e43a..373e51e91ce 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -98,6 +98,7 @@ EXPORT_SYMBOL(pptp_msg_name); static void pptp_expectfn(struct nf_conn *ct, struct nf_conntrack_expect *exp) { + struct net *net = nf_ct_net(ct); typeof(nf_nat_pptp_hook_expectfn) nf_nat_pptp_expectfn; pr_debug("increasing timeouts\n"); @@ -121,7 +122,7 @@ static void pptp_expectfn(struct nf_conn *ct, pr_debug("trying to unexpect other dir: "); nf_ct_dump_tuple(&inv_t); - exp_other = nf_ct_expect_find_get(&inv_t); + exp_other = nf_ct_expect_find_get(net, &inv_t); if (exp_other) { /* delete other expectation. */ pr_debug("found\n"); @@ -134,7 +135,8 @@ static void pptp_expectfn(struct nf_conn *ct, rcu_read_unlock(); } -static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) +static int destroy_sibling_or_exp(struct net *net, + const struct nf_conntrack_tuple *t) { const struct nf_conntrack_tuple_hash *h; struct nf_conntrack_expect *exp; @@ -143,7 +145,7 @@ static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) pr_debug("trying to timeout ct or exp for tuple "); nf_ct_dump_tuple(t); - h = nf_conntrack_find_get(t); + h = nf_conntrack_find_get(net, t); if (h) { sibling = nf_ct_tuplehash_to_ctrack(h); pr_debug("setting timeout of conntrack %p to 0\n", sibling); @@ -154,7 +156,7 @@ static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) nf_ct_put(sibling); return 1; } else { - exp = nf_ct_expect_find_get(t); + exp = nf_ct_expect_find_get(net, t); if (exp) { pr_debug("unexpect_related of expect %p\n", exp); nf_ct_unexpect_related(exp); @@ -168,6 +170,7 @@ static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) /* timeout GRE data connections */ static void pptp_destroy_siblings(struct nf_conn *ct) { + struct net *net = nf_ct_net(ct); const struct nf_conn_help *help = nfct_help(ct); struct nf_conntrack_tuple t; @@ -178,7 +181,7 @@ static void pptp_destroy_siblings(struct nf_conn *ct) t.dst.protonum = IPPROTO_GRE; t.src.u.gre.key = help->help.ct_pptp_info.pns_call_id; t.dst.u.gre.key = help->help.ct_pptp_info.pac_call_id; - if (!destroy_sibling_or_exp(&t)) + if (!destroy_sibling_or_exp(net, &t)) pr_debug("failed to timeout original pns->pac ct/exp\n"); /* try reply (pac->pns) tuple */ @@ -186,7 +189,7 @@ static void pptp_destroy_siblings(struct nf_conn *ct) t.dst.protonum = IPPROTO_GRE; t.src.u.gre.key = help->help.ct_pptp_info.pac_call_id; t.dst.u.gre.key = help->help.ct_pptp_info.pns_call_id; - if (!destroy_sibling_or_exp(&t)) + if (!destroy_sibling_or_exp(net, &t)) pr_debug("failed to timeout reply pac->pns ct/exp\n"); } @@ -594,15 +597,32 @@ static struct nf_conntrack_helper pptp __read_mostly = { .expect_policy = &pptp_exp_policy, }; +static void nf_conntrack_pptp_net_exit(struct net *net) +{ + nf_ct_gre_keymap_flush(net); +} + +static struct pernet_operations nf_conntrack_pptp_net_ops = { + .exit = nf_conntrack_pptp_net_exit, +}; + static int __init nf_conntrack_pptp_init(void) { - return nf_conntrack_helper_register(&pptp); + int rv; + + rv = nf_conntrack_helper_register(&pptp); + if (rv < 0) + return rv; + rv = register_pernet_subsys(&nf_conntrack_pptp_net_ops); + if (rv < 0) + nf_conntrack_helper_unregister(&pptp); + return rv; } static void __exit nf_conntrack_pptp_fini(void) { nf_conntrack_helper_unregister(&pptp); - nf_ct_gre_keymap_flush(); + unregister_pernet_subsys(&nf_conntrack_pptp_net_ops); } module_init(nf_conntrack_pptp_init); diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index a49fc932629..a59a307e685 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -207,6 +207,8 @@ EXPORT_SYMBOL_GPL(nf_conntrack_l3proto_register); void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) { + struct net *net; + BUG_ON(proto->l3proto >= AF_MAX); mutex_lock(&nf_ct_proto_mutex); @@ -219,7 +221,8 @@ void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) synchronize_rcu(); /* Remove all contrack entries for this protocol */ - nf_ct_iterate_cleanup(kill_l3proto, proto); + for_each_net(net) + nf_ct_iterate_cleanup(net, kill_l3proto, proto); } EXPORT_SYMBOL_GPL(nf_conntrack_l3proto_unregister); @@ -316,6 +319,8 @@ EXPORT_SYMBOL_GPL(nf_conntrack_l4proto_register); void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) { + struct net *net; + BUG_ON(l4proto->l3proto >= PF_MAX); mutex_lock(&nf_ct_proto_mutex); @@ -328,7 +333,8 @@ void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) synchronize_rcu(); /* Remove all contrack entries for this protocol */ - nf_ct_iterate_cleanup(kill_l4proto, l4proto); + for_each_net(net) + nf_ct_iterate_cleanup(net, kill_l4proto, l4proto); } EXPORT_SYMBOL_GPL(nf_conntrack_l4proto_unregister); diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index e7866dd3cde..8fcf1762fab 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -418,6 +418,7 @@ static bool dccp_invert_tuple(struct nf_conntrack_tuple *inv, static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff) { + struct net *net = nf_ct_net(ct); struct dccp_hdr _dh, *dh; const char *msg; u_int8_t state; @@ -445,7 +446,7 @@ static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb, return true; out_invalid: - if (LOG_INVALID(IPPROTO_DCCP)) + if (LOG_INVALID(net, IPPROTO_DCCP)) nf_log_packet(nf_ct_l3num(ct), 0, skb, NULL, NULL, NULL, msg); return false; } @@ -461,8 +462,9 @@ static u64 dccp_ack_seq(const struct dccp_hdr *dh) static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, unsigned int hooknum) + u_int8_t pf, unsigned int hooknum) { + struct net *net = nf_ct_net(ct); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); struct dccp_hdr _dh, *dh; u_int8_t type, old_state, new_state; @@ -524,13 +526,13 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.dccp.last_pkt = type; write_unlock_bh(&dccp_lock); - if (LOG_INVALID(IPPROTO_DCCP)) + if (LOG_INVALID(net, IPPROTO_DCCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_dccp: invalid packet ignored "); return NF_ACCEPT; case CT_DCCP_INVALID: write_unlock_bh(&dccp_lock); - if (LOG_INVALID(IPPROTO_DCCP)) + if (LOG_INVALID(net, IPPROTO_DCCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_dccp: invalid state transition "); return -NF_ACCEPT; @@ -545,9 +547,9 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, return NF_ACCEPT; } -static int dccp_error(struct sk_buff *skb, unsigned int dataoff, - enum ip_conntrack_info *ctinfo, int pf, - unsigned int hooknum) +static int dccp_error(struct net *net, struct sk_buff *skb, + unsigned int dataoff, enum ip_conntrack_info *ctinfo, + u_int8_t pf, unsigned int hooknum) { struct dccp_hdr _dh, *dh; unsigned int dccp_len = skb->len - dataoff; @@ -575,7 +577,7 @@ static int dccp_error(struct sk_buff *skb, unsigned int dataoff, } } - if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING && + if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_DCCP, pf)) { msg = "nf_ct_dccp: bad checksum "; @@ -590,7 +592,7 @@ static int dccp_error(struct sk_buff *skb, unsigned int dataoff, return NF_ACCEPT; out_invalid: - if (LOG_INVALID(IPPROTO_DCCP)) + if (LOG_INVALID(net, IPPROTO_DCCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index e31b0e7bd0b..dbe680af85d 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -45,7 +45,7 @@ static int packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_generic_timeout); diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 9bd03967fea..a2cdbcbf64c 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -29,8 +29,11 @@ #include <linux/list.h> #include <linux/seq_file.h> #include <linux/in.h> +#include <linux/netdevice.h> #include <linux/skbuff.h> - +#include <net/dst.h> +#include <net/net_namespace.h> +#include <net/netns/generic.h> #include <net/netfilter/nf_conntrack_l4proto.h> #include <net/netfilter/nf_conntrack_helper.h> #include <net/netfilter/nf_conntrack_core.h> @@ -40,19 +43,23 @@ #define GRE_TIMEOUT (30 * HZ) #define GRE_STREAM_TIMEOUT (180 * HZ) -static DEFINE_RWLOCK(nf_ct_gre_lock); -static LIST_HEAD(gre_keymap_list); +static int proto_gre_net_id; +struct netns_proto_gre { + rwlock_t keymap_lock; + struct list_head keymap_list; +}; -void nf_ct_gre_keymap_flush(void) +void nf_ct_gre_keymap_flush(struct net *net) { + struct netns_proto_gre *net_gre = net_generic(net, proto_gre_net_id); struct nf_ct_gre_keymap *km, *tmp; - write_lock_bh(&nf_ct_gre_lock); - list_for_each_entry_safe(km, tmp, &gre_keymap_list, list) { + write_lock_bh(&net_gre->keymap_lock); + list_for_each_entry_safe(km, tmp, &net_gre->keymap_list, list) { list_del(&km->list); kfree(km); } - write_unlock_bh(&nf_ct_gre_lock); + write_unlock_bh(&net_gre->keymap_lock); } EXPORT_SYMBOL(nf_ct_gre_keymap_flush); @@ -67,19 +74,20 @@ static inline int gre_key_cmpfn(const struct nf_ct_gre_keymap *km, } /* look up the source key for a given tuple */ -static __be16 gre_keymap_lookup(struct nf_conntrack_tuple *t) +static __be16 gre_keymap_lookup(struct net *net, struct nf_conntrack_tuple *t) { + struct netns_proto_gre *net_gre = net_generic(net, proto_gre_net_id); struct nf_ct_gre_keymap *km; __be16 key = 0; - read_lock_bh(&nf_ct_gre_lock); - list_for_each_entry(km, &gre_keymap_list, list) { + read_lock_bh(&net_gre->keymap_lock); + list_for_each_entry(km, &net_gre->keymap_list, list) { if (gre_key_cmpfn(km, t)) { key = km->tuple.src.u.gre.key; break; } } - read_unlock_bh(&nf_ct_gre_lock); + read_unlock_bh(&net_gre->keymap_lock); pr_debug("lookup src key 0x%x for ", key); nf_ct_dump_tuple(t); @@ -91,20 +99,22 @@ static __be16 gre_keymap_lookup(struct nf_conntrack_tuple *t) int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, struct nf_conntrack_tuple *t) { + struct net *net = nf_ct_net(ct); + struct netns_proto_gre *net_gre = net_generic(net, proto_gre_net_id); struct nf_conn_help *help = nfct_help(ct); struct nf_ct_gre_keymap **kmp, *km; kmp = &help->help.ct_pptp_info.keymap[dir]; if (*kmp) { /* check whether it's a retransmission */ - read_lock_bh(&nf_ct_gre_lock); - list_for_each_entry(km, &gre_keymap_list, list) { + read_lock_bh(&net_gre->keymap_lock); + list_for_each_entry(km, &net_gre->keymap_list, list) { if (gre_key_cmpfn(km, t) && km == *kmp) { - read_unlock_bh(&nf_ct_gre_lock); + read_unlock_bh(&net_gre->keymap_lock); return 0; } } - read_unlock_bh(&nf_ct_gre_lock); + read_unlock_bh(&net_gre->keymap_lock); pr_debug("trying to override keymap_%s for ct %p\n", dir == IP_CT_DIR_REPLY ? "reply" : "orig", ct); return -EEXIST; @@ -119,9 +129,9 @@ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, pr_debug("adding new entry %p: ", km); nf_ct_dump_tuple(&km->tuple); - write_lock_bh(&nf_ct_gre_lock); - list_add_tail(&km->list, &gre_keymap_list); - write_unlock_bh(&nf_ct_gre_lock); + write_lock_bh(&net_gre->keymap_lock); + list_add_tail(&km->list, &net_gre->keymap_list); + write_unlock_bh(&net_gre->keymap_lock); return 0; } @@ -130,12 +140,14 @@ EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_add); /* destroy the keymap entries associated with specified master ct */ void nf_ct_gre_keymap_destroy(struct nf_conn *ct) { + struct net *net = nf_ct_net(ct); + struct netns_proto_gre *net_gre = net_generic(net, proto_gre_net_id); struct nf_conn_help *help = nfct_help(ct); enum ip_conntrack_dir dir; pr_debug("entering for ct %p\n", ct); - write_lock_bh(&nf_ct_gre_lock); + write_lock_bh(&net_gre->keymap_lock); for (dir = IP_CT_DIR_ORIGINAL; dir < IP_CT_DIR_MAX; dir++) { if (help->help.ct_pptp_info.keymap[dir]) { pr_debug("removing %p from list\n", @@ -145,7 +157,7 @@ void nf_ct_gre_keymap_destroy(struct nf_conn *ct) help->help.ct_pptp_info.keymap[dir] = NULL; } } - write_unlock_bh(&nf_ct_gre_lock); + write_unlock_bh(&net_gre->keymap_lock); } EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy); @@ -164,6 +176,7 @@ static bool gre_invert_tuple(struct nf_conntrack_tuple *tuple, static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, struct nf_conntrack_tuple *tuple) { + struct net *net = dev_net(skb->dev ? skb->dev : skb->dst->dev); const struct gre_hdr_pptp *pgrehdr; struct gre_hdr_pptp _pgrehdr; __be16 srckey; @@ -190,7 +203,7 @@ static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, } tuple->dst.u.gre.key = pgrehdr->call_id; - srckey = gre_keymap_lookup(tuple); + srckey = gre_keymap_lookup(net, tuple); tuple->src.u.gre.key = srckey; return true; @@ -219,7 +232,7 @@ static int gre_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { /* If we've seen traffic both ways, this is a GRE connection. @@ -229,7 +242,7 @@ static int gre_packet(struct nf_conn *ct, ct->proto.gre.stream_timeout); /* Also, more likely to be important, and not a probe. */ set_bit(IPS_ASSURED_BIT, &ct->status); - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_STATUS, ct); } else nf_ct_refresh_acct(ct, ctinfo, skb, ct->proto.gre.timeout); @@ -285,15 +298,53 @@ static struct nf_conntrack_l4proto nf_conntrack_l4proto_gre4 __read_mostly = { #endif }; +static int proto_gre_net_init(struct net *net) +{ + struct netns_proto_gre *net_gre; + int rv; + + net_gre = kmalloc(sizeof(struct netns_proto_gre), GFP_KERNEL); + if (!net_gre) + return -ENOMEM; + rwlock_init(&net_gre->keymap_lock); + INIT_LIST_HEAD(&net_gre->keymap_list); + + rv = net_assign_generic(net, proto_gre_net_id, net_gre); + if (rv < 0) + kfree(net_gre); + return rv; +} + +static void proto_gre_net_exit(struct net *net) +{ + struct netns_proto_gre *net_gre = net_generic(net, proto_gre_net_id); + + nf_ct_gre_keymap_flush(net); + kfree(net_gre); +} + +static struct pernet_operations proto_gre_net_ops = { + .init = proto_gre_net_init, + .exit = proto_gre_net_exit, +}; + static int __init nf_ct_proto_gre_init(void) { - return nf_conntrack_l4proto_register(&nf_conntrack_l4proto_gre4); + int rv; + + rv = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_gre4); + if (rv < 0) + return rv; + rv = register_pernet_gen_device(&proto_gre_net_id, &proto_gre_net_ops); + if (rv < 0) + nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_gre4); + return rv; } static void nf_ct_proto_gre_fini(void) { nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_gre4); - nf_ct_gre_keymap_flush(); + unregister_pernet_gen_device(proto_gre_net_id, &proto_gre_net_ops); } module_init(nf_ct_proto_gre_init); diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 30aa5b94a77..ae8c2609e23 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -287,7 +287,7 @@ static int sctp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { enum sctp_conntrack new_state, old_state; @@ -369,7 +369,7 @@ static int sctp_packet(struct nf_conn *ct, ct->proto.sctp.state = new_state; if (old_state != new_state) - nf_conntrack_event_cache(IPCT_PROTOINFO, skb); + nf_conntrack_event_cache(IPCT_PROTOINFO, ct); } write_unlock_bh(&sctp_lock); @@ -380,7 +380,7 @@ static int sctp_packet(struct nf_conn *ct, new_state == SCTP_CONNTRACK_ESTABLISHED) { pr_debug("Setting assured bit\n"); set_bit(IPS_ASSURED_BIT, &ct->status); - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_STATUS, ct); } return NF_ACCEPT; diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 6f61261888e..f947ec41e39 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -486,8 +486,9 @@ static bool tcp_in_window(const struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, const struct tcphdr *tcph, - int pf) + u_int8_t pf) { + struct net *net = nf_ct_net(ct); struct ip_ct_tcp_state *sender = &state->seen[dir]; struct ip_ct_tcp_state *receiver = &state->seen[!dir]; const struct nf_conntrack_tuple *tuple = &ct->tuplehash[dir].tuple; @@ -668,7 +669,7 @@ static bool tcp_in_window(const struct nf_conn *ct, if (sender->flags & IP_CT_TCP_FLAG_BE_LIBERAL || nf_ct_tcp_be_liberal) res = true; - if (!res && LOG_INVALID(IPPROTO_TCP)) + if (!res && LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: %s ", before(seq, sender->td_maxend + 1) ? @@ -746,10 +747,11 @@ static const u8 tcp_valid_flags[(TH_FIN|TH_SYN|TH_RST|TH_ACK|TH_URG) + 1] = }; /* Protect conntrack agaist broken packets. Code taken from ipt_unclean.c. */ -static int tcp_error(struct sk_buff *skb, +static int tcp_error(struct net *net, + struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info *ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { const struct tcphdr *th; @@ -760,7 +762,7 @@ static int tcp_error(struct sk_buff *skb, /* Smaller that minimal TCP header? */ th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); if (th == NULL) { - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: short packet "); return -NF_ACCEPT; @@ -768,7 +770,7 @@ static int tcp_error(struct sk_buff *skb, /* Not whole TCP header or malformed packet */ if (th->doff*4 < sizeof(struct tcphdr) || tcplen < th->doff*4) { - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: truncated/malformed packet "); return -NF_ACCEPT; @@ -779,9 +781,9 @@ static int tcp_error(struct sk_buff *skb, * because the checksum is assumed to be correct. */ /* FIXME: Source route IP option packets --RR */ - if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING && + if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum(skb, hooknum, dataoff, IPPROTO_TCP, pf)) { - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: bad TCP checksum "); return -NF_ACCEPT; @@ -790,7 +792,7 @@ static int tcp_error(struct sk_buff *skb, /* Check TCP flags. */ tcpflags = (((u_int8_t *)th)[13] & ~(TH_ECE|TH_CWR|TH_PUSH)); if (!tcp_valid_flags[tcpflags]) { - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: invalid TCP flag combination "); return -NF_ACCEPT; @@ -804,9 +806,10 @@ static int tcp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { + struct net *net = nf_ct_net(ct); struct nf_conntrack_tuple *tuple; enum tcp_conntrack new_state, old_state; enum ip_conntrack_dir dir; @@ -885,7 +888,7 @@ static int tcp_packet(struct nf_conn *ct, * thus initiate a clean new session. */ write_unlock_bh(&tcp_lock); - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: killing out of sync session "); nf_ct_kill(ct); @@ -898,7 +901,7 @@ static int tcp_packet(struct nf_conn *ct, segment_seq_plus_len(ntohl(th->seq), skb->len, dataoff, th); write_unlock_bh(&tcp_lock); - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: invalid packet ignored "); return NF_ACCEPT; @@ -907,7 +910,7 @@ static int tcp_packet(struct nf_conn *ct, pr_debug("nf_ct_tcp: Invalid dir=%i index=%u ostate=%u\n", dir, get_conntrack_index(th), old_state); write_unlock_bh(&tcp_lock); - if (LOG_INVALID(IPPROTO_TCP)) + if (LOG_INVALID(net, IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: invalid state "); return -NF_ACCEPT; @@ -968,9 +971,9 @@ static int tcp_packet(struct nf_conn *ct, timeout = tcp_timeouts[new_state]; write_unlock_bh(&tcp_lock); - nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, ct); if (new_state != old_state) - nf_conntrack_event_cache(IPCT_PROTOINFO, skb); + nf_conntrack_event_cache(IPCT_PROTOINFO, ct); if (!test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) { /* If only reply is a RST, we can consider ourselves not to @@ -989,7 +992,7 @@ static int tcp_packet(struct nf_conn *ct, after SYN_RECV or a valid answer for a picked up connection. */ set_bit(IPS_ASSURED_BIT, &ct->status); - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_STATUS, ct); } nf_ct_refresh_acct(ct, ctinfo, skb, timeout); diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index 8b21762e65d..7c2ca48698b 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -66,7 +66,7 @@ static int udp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { /* If we've seen traffic both ways, this is some kind of UDP @@ -75,7 +75,7 @@ static int udp_packet(struct nf_conn *ct, nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_udp_timeout_stream); /* Also, more likely to be important, and not a probe */ if (!test_and_set_bit(IPS_ASSURED_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_STATUS, ct); } else nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_udp_timeout); @@ -89,9 +89,9 @@ static bool udp_new(struct nf_conn *ct, const struct sk_buff *skb, return true; } -static int udp_error(struct sk_buff *skb, unsigned int dataoff, +static int udp_error(struct net *net, struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info *ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { unsigned int udplen = skb->len - dataoff; @@ -101,7 +101,7 @@ static int udp_error(struct sk_buff *skb, unsigned int dataoff, /* Header is too small? */ hdr = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hdr == NULL) { - if (LOG_INVALID(IPPROTO_UDP)) + if (LOG_INVALID(net, IPPROTO_UDP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udp: short packet "); return -NF_ACCEPT; @@ -109,7 +109,7 @@ static int udp_error(struct sk_buff *skb, unsigned int dataoff, /* Truncated/malformed packets */ if (ntohs(hdr->len) > udplen || ntohs(hdr->len) < sizeof(*hdr)) { - if (LOG_INVALID(IPPROTO_UDP)) + if (LOG_INVALID(net, IPPROTO_UDP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udp: truncated/malformed packet "); return -NF_ACCEPT; @@ -123,9 +123,9 @@ static int udp_error(struct sk_buff *skb, unsigned int dataoff, * We skip checking packets on the outgoing path * because the checksum is assumed to be correct. * FIXME: Source route IP option packets --RR */ - if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING && + if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum(skb, hooknum, dataoff, IPPROTO_UDP, pf)) { - if (LOG_INVALID(IPPROTO_UDP)) + if (LOG_INVALID(net, IPPROTO_UDP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udp: bad UDP checksum "); return -NF_ACCEPT; diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c index 1fa62f3c24f..d22d839e4f9 100644 --- a/net/netfilter/nf_conntrack_proto_udplite.c +++ b/net/netfilter/nf_conntrack_proto_udplite.c @@ -65,7 +65,7 @@ static int udplite_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { /* If we've seen traffic both ways, this is some kind of UDP @@ -75,7 +75,7 @@ static int udplite_packet(struct nf_conn *ct, nf_ct_udplite_timeout_stream); /* Also, more likely to be important, and not a probe */ if (!test_and_set_bit(IPS_ASSURED_BIT, &ct->status)) - nf_conntrack_event_cache(IPCT_STATUS, skb); + nf_conntrack_event_cache(IPCT_STATUS, ct); } else nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_udplite_timeout); @@ -89,9 +89,11 @@ static bool udplite_new(struct nf_conn *ct, const struct sk_buff *skb, return true; } -static int udplite_error(struct sk_buff *skb, unsigned int dataoff, +static int udplite_error(struct net *net, + struct sk_buff *skb, + unsigned int dataoff, enum ip_conntrack_info *ctinfo, - int pf, + u_int8_t pf, unsigned int hooknum) { unsigned int udplen = skb->len - dataoff; @@ -102,7 +104,7 @@ static int udplite_error(struct sk_buff *skb, unsigned int dataoff, /* Header is too small? */ hdr = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hdr == NULL) { - if (LOG_INVALID(IPPROTO_UDPLITE)) + if (LOG_INVALID(net, IPPROTO_UDPLITE)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udplite: short packet "); return -NF_ACCEPT; @@ -112,7 +114,7 @@ static int udplite_error(struct sk_buff *skb, unsigned int dataoff, if (cscov == 0) cscov = udplen; else if (cscov < sizeof(*hdr) || cscov > udplen) { - if (LOG_INVALID(IPPROTO_UDPLITE)) + if (LOG_INVALID(net, IPPROTO_UDPLITE)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udplite: invalid checksum coverage "); return -NF_ACCEPT; @@ -120,17 +122,17 @@ static int udplite_error(struct sk_buff *skb, unsigned int dataoff, /* UDPLITE mandates checksums */ if (!hdr->check) { - if (LOG_INVALID(IPPROTO_UDPLITE)) + if (LOG_INVALID(net, IPPROTO_UDPLITE)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udplite: checksum missing "); return -NF_ACCEPT; } /* Checksum invalid? Ignore. */ - if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING && + if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_UDP, pf)) { - if (LOG_INVALID(IPPROTO_UDPLITE)) + if (LOG_INVALID(net, IPPROTO_UDPLITE)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_udplite: bad UDPLite checksum "); return -NF_ACCEPT; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 1fa306be60f..6813f1c8863 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -736,6 +736,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp; enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct net *net = nf_ct_net(ct); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); union nf_inet_addr *saddr; struct nf_conntrack_tuple tuple; @@ -775,7 +776,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, rcu_read_lock(); do { - exp = __nf_ct_expect_find(&tuple); + exp = __nf_ct_expect_find(net, &tuple); if (!exp || exp->master == ct || nfct_help(exp->master)->helper != nfct_help(ct)->helper || diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 8509db14670..98106d4e89f 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -40,18 +40,20 @@ print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, EXPORT_SYMBOL_GPL(print_tuple); struct ct_iter_state { + struct seq_net_private p; unsigned int bucket; }; static struct hlist_node *ct_get_first(struct seq_file *seq) { + struct net *net = seq_file_net(seq); struct ct_iter_state *st = seq->private; struct hlist_node *n; for (st->bucket = 0; st->bucket < nf_conntrack_htable_size; st->bucket++) { - n = rcu_dereference(nf_conntrack_hash[st->bucket].first); + n = rcu_dereference(net->ct.hash[st->bucket].first); if (n) return n; } @@ -61,13 +63,14 @@ static struct hlist_node *ct_get_first(struct seq_file *seq) static struct hlist_node *ct_get_next(struct seq_file *seq, struct hlist_node *head) { + struct net *net = seq_file_net(seq); struct ct_iter_state *st = seq->private; head = rcu_dereference(head->next); while (head == NULL) { if (++st->bucket >= nf_conntrack_htable_size) return NULL; - head = rcu_dereference(nf_conntrack_hash[st->bucket].first); + head = rcu_dereference(net->ct.hash[st->bucket].first); } return head; } @@ -177,7 +180,7 @@ static const struct seq_operations ct_seq_ops = { static int ct_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &ct_seq_ops, + return seq_open_net(inode, file, &ct_seq_ops, sizeof(struct ct_iter_state)); } @@ -186,11 +189,12 @@ static const struct file_operations ct_file_ops = { .open = ct_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) { + struct net *net = seq_file_net(seq); int cpu; if (*pos == 0) @@ -200,7 +204,7 @@ static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) if (!cpu_possible(cpu)) continue; *pos = cpu + 1; - return &per_cpu(nf_conntrack_stat, cpu); + return per_cpu_ptr(net->ct.stat, cpu); } return NULL; @@ -208,13 +212,14 @@ static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) static void *ct_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) { + struct net *net = seq_file_net(seq); int cpu; for (cpu = *pos; cpu < NR_CPUS; ++cpu) { if (!cpu_possible(cpu)) continue; *pos = cpu + 1; - return &per_cpu(nf_conntrack_stat, cpu); + return per_cpu_ptr(net->ct.stat, cpu); } return NULL; @@ -226,7 +231,8 @@ static void ct_cpu_seq_stop(struct seq_file *seq, void *v) static int ct_cpu_seq_show(struct seq_file *seq, void *v) { - unsigned int nr_conntracks = atomic_read(&nf_conntrack_count); + struct net *net = seq_file_net(seq); + unsigned int nr_conntracks = atomic_read(&net->ct.count); const struct ip_conntrack_stat *st = v; if (v == SEQ_START_TOKEN) { @@ -266,7 +272,8 @@ static const struct seq_operations ct_cpu_seq_ops = { static int ct_cpu_seq_open(struct inode *inode, struct file *file) { - return seq_open(file, &ct_cpu_seq_ops); + return seq_open_net(inode, file, &ct_cpu_seq_ops, + sizeof(struct seq_net_private)); } static const struct file_operations ct_cpu_seq_fops = { @@ -274,56 +281,52 @@ static const struct file_operations ct_cpu_seq_fops = { .open = ct_cpu_seq_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = seq_release_net, }; -static int nf_conntrack_standalone_init_proc(void) +static int nf_conntrack_standalone_init_proc(struct net *net) { struct proc_dir_entry *pde; - pde = proc_net_fops_create(&init_net, "nf_conntrack", 0440, &ct_file_ops); + pde = proc_net_fops_create(net, "nf_conntrack", 0440, &ct_file_ops); if (!pde) goto out_nf_conntrack; - pde = proc_create("nf_conntrack", S_IRUGO, init_net.proc_net_stat, + pde = proc_create("nf_conntrack", S_IRUGO, net->proc_net_stat, &ct_cpu_seq_fops); if (!pde) goto out_stat_nf_conntrack; return 0; out_stat_nf_conntrack: - proc_net_remove(&init_net, "nf_conntrack"); + proc_net_remove(net, "nf_conntrack"); out_nf_conntrack: return -ENOMEM; } -static void nf_conntrack_standalone_fini_proc(void) +static void nf_conntrack_standalone_fini_proc(struct net *net) { - remove_proc_entry("nf_conntrack", init_net.proc_net_stat); - proc_net_remove(&init_net, "nf_conntrack"); + remove_proc_entry("nf_conntrack", net->proc_net_stat); + proc_net_remove(net, "nf_conntrack"); } #else -static int nf_conntrack_standalone_init_proc(void) +static int nf_conntrack_standalone_init_proc(struct net *net) { return 0; } -static void nf_conntrack_standalone_fini_proc(void) +static void nf_conntrack_standalone_fini_proc(struct net *net) { } #endif /* CONFIG_PROC_FS */ /* Sysctl support */ -int nf_conntrack_checksum __read_mostly = 1; -EXPORT_SYMBOL_GPL(nf_conntrack_checksum); - #ifdef CONFIG_SYSCTL /* Log invalid packets of a given protocol */ static int log_invalid_proto_min = 0; static int log_invalid_proto_max = 255; -static struct ctl_table_header *nf_ct_sysctl_header; static struct ctl_table_header *nf_ct_netfilter_header; static ctl_table nf_ct_sysctl_table[] = { @@ -338,7 +341,7 @@ static ctl_table nf_ct_sysctl_table[] = { { .ctl_name = NET_NF_CONNTRACK_COUNT, .procname = "nf_conntrack_count", - .data = &nf_conntrack_count, + .data = &init_net.ct.count, .maxlen = sizeof(int), .mode = 0444, .proc_handler = &proc_dointvec, @@ -354,7 +357,7 @@ static ctl_table nf_ct_sysctl_table[] = { { .ctl_name = NET_NF_CONNTRACK_CHECKSUM, .procname = "nf_conntrack_checksum", - .data = &nf_conntrack_checksum, + .data = &init_net.ct.sysctl_checksum, .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = &proc_dointvec, @@ -362,7 +365,7 @@ static ctl_table nf_ct_sysctl_table[] = { { .ctl_name = NET_NF_CONNTRACK_LOG_INVALID, .procname = "nf_conntrack_log_invalid", - .data = &nf_ct_log_invalid, + .data = &init_net.ct.sysctl_log_invalid, .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = &proc_dointvec_minmax, @@ -400,74 +403,109 @@ static struct ctl_path nf_ct_path[] = { { } }; -EXPORT_SYMBOL_GPL(nf_ct_log_invalid); - -static int nf_conntrack_standalone_init_sysctl(void) +static int nf_conntrack_standalone_init_sysctl(struct net *net) { - nf_ct_netfilter_header = - register_sysctl_paths(nf_ct_path, nf_ct_netfilter_table); - if (!nf_ct_netfilter_header) - goto out; - - nf_ct_sysctl_header = - register_sysctl_paths(nf_net_netfilter_sysctl_path, - nf_ct_sysctl_table); - if (!nf_ct_sysctl_header) + struct ctl_table *table; + + if (net_eq(net, &init_net)) { + nf_ct_netfilter_header = + register_sysctl_paths(nf_ct_path, nf_ct_netfilter_table); + if (!nf_ct_netfilter_header) + goto out; + } + + table = kmemdup(nf_ct_sysctl_table, sizeof(nf_ct_sysctl_table), + GFP_KERNEL); + if (!table) + goto out_kmemdup; + + table[1].data = &net->ct.count; + table[3].data = &net->ct.sysctl_checksum; + table[4].data = &net->ct.sysctl_log_invalid; + + net->ct.sysctl_header = register_net_sysctl_table(net, + nf_net_netfilter_sysctl_path, table); + if (!net->ct.sysctl_header) goto out_unregister_netfilter; return 0; out_unregister_netfilter: - unregister_sysctl_table(nf_ct_netfilter_header); + kfree(table); +out_kmemdup: + if (net_eq(net, &init_net)) + unregister_sysctl_table(nf_ct_netfilter_header); out: printk("nf_conntrack: can't register to sysctl.\n"); return -ENOMEM; } -static void nf_conntrack_standalone_fini_sysctl(void) +static void nf_conntrack_standalone_fini_sysctl(struct net *net) { - unregister_sysctl_table(nf_ct_netfilter_header); - unregister_sysctl_table(nf_ct_sysctl_header); + struct ctl_table *table; + + if (net_eq(net, &init_net)) + unregister_sysctl_table(nf_ct_netfilter_header); + table = net->ct.sysctl_header->ctl_table_arg; + unregister_net_sysctl_table(net->ct.sysctl_header); + kfree(table); } #else -static int nf_conntrack_standalone_init_sysctl(void) +static int nf_conntrack_standalone_init_sysctl(struct net *net) { return 0; } -static void nf_conntrack_standalone_fini_sysctl(void) +static void nf_conntrack_standalone_fini_sysctl(struct net *net) { } #endif /* CONFIG_SYSCTL */ -static int __init nf_conntrack_standalone_init(void) +static int nf_conntrack_net_init(struct net *net) { int ret; - ret = nf_conntrack_init(); + ret = nf_conntrack_init(net); if (ret < 0) - goto out; - ret = nf_conntrack_standalone_init_proc(); + goto out_init; + ret = nf_conntrack_standalone_init_proc(net); if (ret < 0) goto out_proc; - ret = nf_conntrack_standalone_init_sysctl(); + net->ct.sysctl_checksum = 1; + net->ct.sysctl_log_invalid = 0; + ret = nf_conntrack_standalone_init_sysctl(net); if (ret < 0) goto out_sysctl; return 0; out_sysctl: - nf_conntrack_standalone_fini_proc(); + nf_conntrack_standalone_fini_proc(net); out_proc: - nf_conntrack_cleanup(); -out: + nf_conntrack_cleanup(net); +out_init: return ret; } +static void nf_conntrack_net_exit(struct net *net) +{ + nf_conntrack_standalone_fini_sysctl(net); + nf_conntrack_standalone_fini_proc(net); + nf_conntrack_cleanup(net); +} + +static struct pernet_operations nf_conntrack_net_ops = { + .init = nf_conntrack_net_init, + .exit = nf_conntrack_net_exit, +}; + +static int __init nf_conntrack_standalone_init(void) +{ + return register_pernet_subsys(&nf_conntrack_net_ops); +} + static void __exit nf_conntrack_standalone_fini(void) { - nf_conntrack_standalone_fini_sysctl(); - nf_conntrack_standalone_fini_proc(); - nf_conntrack_cleanup(); + unregister_pernet_subsys(&nf_conntrack_net_ops); } module_init(nf_conntrack_standalone_init); diff --git a/net/netfilter/nf_internals.h b/net/netfilter/nf_internals.h index 196269c1e58..bf6609978af 100644 --- a/net/netfilter/nf_internals.h +++ b/net/netfilter/nf_internals.h @@ -15,7 +15,7 @@ /* core.c */ extern unsigned int nf_iterate(struct list_head *head, struct sk_buff *skb, - int hook, + unsigned int hook, const struct net_device *indev, const struct net_device *outdev, struct list_head **i, @@ -25,7 +25,7 @@ extern unsigned int nf_iterate(struct list_head *head, /* nf_queue.c */ extern int nf_queue(struct sk_buff *skb, struct list_head *elem, - int pf, unsigned int hook, + u_int8_t pf, unsigned int hook, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *), diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 9fda6ee95a3..fa8ae5d2659 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -15,16 +15,16 @@ #define NF_LOG_PREFIXLEN 128 -static const struct nf_logger *nf_loggers[NPROTO] __read_mostly; +static const struct nf_logger *nf_loggers[NFPROTO_NUMPROTO] __read_mostly; static DEFINE_MUTEX(nf_log_mutex); /* return EBUSY if somebody else is registered, EEXIST if the same logger * is registred, 0 on success. */ -int nf_log_register(int pf, const struct nf_logger *logger) +int nf_log_register(u_int8_t pf, const struct nf_logger *logger) { int ret; - if (pf >= NPROTO) + if (pf >= ARRAY_SIZE(nf_loggers)) return -EINVAL; /* Any setup of logging members must be done before @@ -45,9 +45,9 @@ int nf_log_register(int pf, const struct nf_logger *logger) } EXPORT_SYMBOL(nf_log_register); -void nf_log_unregister_pf(int pf) +void nf_log_unregister_pf(u_int8_t pf) { - if (pf >= NPROTO) + if (pf >= ARRAY_SIZE(nf_loggers)) return; mutex_lock(&nf_log_mutex); rcu_assign_pointer(nf_loggers[pf], NULL); @@ -63,7 +63,7 @@ void nf_log_unregister(const struct nf_logger *logger) int i; mutex_lock(&nf_log_mutex); - for (i = 0; i < NPROTO; i++) { + for (i = 0; i < ARRAY_SIZE(nf_loggers); i++) { if (nf_loggers[i] == logger) rcu_assign_pointer(nf_loggers[i], NULL); } @@ -73,7 +73,7 @@ void nf_log_unregister(const struct nf_logger *logger) } EXPORT_SYMBOL(nf_log_unregister); -void nf_log_packet(int pf, +void nf_log_packet(u_int8_t pf, unsigned int hooknum, const struct sk_buff *skb, const struct net_device *in, @@ -103,7 +103,7 @@ static void *seq_start(struct seq_file *seq, loff_t *pos) { rcu_read_lock(); - if (*pos >= NPROTO) + if (*pos >= ARRAY_SIZE(nf_loggers)) return NULL; return pos; @@ -113,7 +113,7 @@ static void *seq_next(struct seq_file *s, void *v, loff_t *pos) { (*pos)++; - if (*pos >= NPROTO) + if (*pos >= ARRAY_SIZE(nf_loggers)) return NULL; return pos; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 582ec3efc8a..4f2310c93e0 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -16,17 +16,17 @@ * long term mutex. The handler must provide an an outfn() to accept packets * for queueing and must reinject all packets it receives, no matter what. */ -static const struct nf_queue_handler *queue_handler[NPROTO]; +static const struct nf_queue_handler *queue_handler[NFPROTO_NUMPROTO] __read_mostly; static DEFINE_MUTEX(queue_handler_mutex); /* return EBUSY when somebody else is registered, return EEXIST if the * same handler is registered, return 0 in case of success. */ -int nf_register_queue_handler(int pf, const struct nf_queue_handler *qh) +int nf_register_queue_handler(u_int8_t pf, const struct nf_queue_handler *qh) { int ret; - if (pf >= NPROTO) + if (pf >= ARRAY_SIZE(queue_handler)) return -EINVAL; mutex_lock(&queue_handler_mutex); @@ -45,9 +45,9 @@ int nf_register_queue_handler(int pf, const struct nf_queue_handler *qh) EXPORT_SYMBOL(nf_register_queue_handler); /* The caller must flush their queue before this */ -int nf_unregister_queue_handler(int pf, const struct nf_queue_handler *qh) +int nf_unregister_queue_handler(u_int8_t pf, const struct nf_queue_handler *qh) { - if (pf >= NPROTO) + if (pf >= ARRAY_SIZE(queue_handler)) return -EINVAL; mutex_lock(&queue_handler_mutex); @@ -67,10 +67,10 @@ EXPORT_SYMBOL(nf_unregister_queue_handler); void nf_unregister_queue_handlers(const struct nf_queue_handler *qh) { - int pf; + u_int8_t pf; mutex_lock(&queue_handler_mutex); - for (pf = 0; pf < NPROTO; pf++) { + for (pf = 0; pf < ARRAY_SIZE(queue_handler); pf++) { if (queue_handler[pf] == qh) rcu_assign_pointer(queue_handler[pf], NULL); } @@ -107,7 +107,7 @@ static void nf_queue_entry_release_refs(struct nf_queue_entry *entry) */ static int __nf_queue(struct sk_buff *skb, struct list_head *elem, - int pf, unsigned int hook, + u_int8_t pf, unsigned int hook, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *), @@ -191,7 +191,7 @@ err: int nf_queue(struct sk_buff *skb, struct list_head *elem, - int pf, unsigned int hook, + u_int8_t pf, unsigned int hook, struct net_device *indev, struct net_device *outdev, int (*okfn)(struct sk_buff *), @@ -285,7 +285,7 @@ EXPORT_SYMBOL(nf_reinject); #ifdef CONFIG_PROC_FS static void *seq_start(struct seq_file *seq, loff_t *pos) { - if (*pos >= NPROTO) + if (*pos >= ARRAY_SIZE(queue_handler)) return NULL; return pos; @@ -295,7 +295,7 @@ static void *seq_next(struct seq_file *s, void *v, loff_t *pos) { (*pos)++; - if (*pos >= NPROTO) + if (*pos >= ARRAY_SIZE(queue_handler)) return NULL; return pos; diff --git a/net/netfilter/nf_sockopt.c b/net/netfilter/nf_sockopt.c index 01489681fa9..8ab829f8657 100644 --- a/net/netfilter/nf_sockopt.c +++ b/net/netfilter/nf_sockopt.c @@ -60,14 +60,11 @@ void nf_unregister_sockopt(struct nf_sockopt_ops *reg) } EXPORT_SYMBOL(nf_unregister_sockopt); -static struct nf_sockopt_ops *nf_sockopt_find(struct sock *sk, int pf, +static struct nf_sockopt_ops *nf_sockopt_find(struct sock *sk, u_int8_t pf, int val, int get) { struct nf_sockopt_ops *ops; - if (!net_eq(sock_net(sk), &init_net)) - return ERR_PTR(-ENOPROTOOPT); - if (mutex_lock_interruptible(&nf_sockopt_mutex) != 0) return ERR_PTR(-EINTR); @@ -96,7 +93,7 @@ out: } /* Call get/setsockopt() */ -static int nf_sockopt(struct sock *sk, int pf, int val, +static int nf_sockopt(struct sock *sk, u_int8_t pf, int val, char __user *opt, int *len, int get) { struct nf_sockopt_ops *ops; @@ -115,21 +112,22 @@ static int nf_sockopt(struct sock *sk, int pf, int val, return ret; } -int nf_setsockopt(struct sock *sk, int pf, int val, char __user *opt, +int nf_setsockopt(struct sock *sk, u_int8_t pf, int val, char __user *opt, int len) { return nf_sockopt(sk, pf, val, opt, &len, 0); } EXPORT_SYMBOL(nf_setsockopt); -int nf_getsockopt(struct sock *sk, int pf, int val, char __user *opt, int *len) +int nf_getsockopt(struct sock *sk, u_int8_t pf, int val, char __user *opt, + int *len) { return nf_sockopt(sk, pf, val, opt, len, 1); } EXPORT_SYMBOL(nf_getsockopt); #ifdef CONFIG_COMPAT -static int compat_nf_sockopt(struct sock *sk, int pf, int val, +static int compat_nf_sockopt(struct sock *sk, u_int8_t pf, int val, char __user *opt, int *len, int get) { struct nf_sockopt_ops *ops; @@ -155,14 +153,14 @@ static int compat_nf_sockopt(struct sock *sk, int pf, int val, return ret; } -int compat_nf_setsockopt(struct sock *sk, int pf, +int compat_nf_setsockopt(struct sock *sk, u_int8_t pf, int val, char __user *opt, int len) { return compat_nf_sockopt(sk, pf, val, opt, &len, 0); } EXPORT_SYMBOL(compat_nf_setsockopt); -int compat_nf_getsockopt(struct sock *sk, int pf, +int compat_nf_getsockopt(struct sock *sk, u_int8_t pf, int val, char __user *opt, int *len) { return compat_nf_sockopt(sk, pf, val, opt, len, 1); diff --git a/net/netfilter/nf_tproxy_core.c b/net/netfilter/nf_tproxy_core.c new file mode 100644 index 00000000000..fe34f4bf74c --- /dev/null +++ b/net/netfilter/nf_tproxy_core.c @@ -0,0 +1,96 @@ +/* + * Transparent proxy support for Linux/iptables + * + * Copyright (c) 2006-2007 BalaBit IT Ltd. + * Author: Balazs Scheidler, Krisztian Kovacs + * + * 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/version.h> +#include <linux/module.h> + +#include <linux/net.h> +#include <linux/if.h> +#include <linux/netdevice.h> +#include <net/udp.h> +#include <net/netfilter/nf_tproxy_core.h> + +struct sock * +nf_tproxy_get_sock_v4(struct net *net, const u8 protocol, + const __be32 saddr, const __be32 daddr, + const __be16 sport, const __be16 dport, + const struct net_device *in, bool listening_only) +{ + struct sock *sk; + + /* look up socket */ + switch (protocol) { + case IPPROTO_TCP: + if (listening_only) + sk = __inet_lookup_listener(net, &tcp_hashinfo, + daddr, ntohs(dport), + in->ifindex); + else + sk = __inet_lookup(net, &tcp_hashinfo, + saddr, sport, daddr, dport, + in->ifindex); + break; + case IPPROTO_UDP: + sk = udp4_lib_lookup(net, saddr, sport, daddr, dport, + in->ifindex); + break; + default: + WARN_ON(1); + sk = NULL; + } + + pr_debug("tproxy socket lookup: proto %u %08x:%u -> %08x:%u, listener only: %d, sock %p\n", + protocol, ntohl(saddr), ntohs(sport), ntohl(daddr), ntohs(dport), listening_only, sk); + + return sk; +} +EXPORT_SYMBOL_GPL(nf_tproxy_get_sock_v4); + +static void +nf_tproxy_destructor(struct sk_buff *skb) +{ + struct sock *sk = skb->sk; + + skb->sk = NULL; + skb->destructor = NULL; + + if (sk) + nf_tproxy_put_sock(sk); +} + +/* consumes sk */ +int +nf_tproxy_assign_sock(struct sk_buff *skb, struct sock *sk) +{ + if (inet_sk(sk)->transparent) { + skb->sk = sk; + skb->destructor = nf_tproxy_destructor; + return 1; + } else + nf_tproxy_put_sock(sk); + + return 0; +} +EXPORT_SYMBOL_GPL(nf_tproxy_assign_sock); + +static int __init nf_tproxy_init(void) +{ + pr_info("NF_TPROXY: Transparent proxy support initialized, version 4.1.0\n"); + pr_info("NF_TPROXY: Copyright (c) 2006-2007 BalaBit IT Ltd.\n"); + return 0; +} + +module_init(nf_tproxy_init); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Krisztian Kovacs"); +MODULE_DESCRIPTION("Transparent proxy support core routines"); diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 9a35b57ab76..41e0105d382 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -359,7 +359,7 @@ static inline int __build_packet_message(struct nfulnl_instance *inst, const struct sk_buff *skb, unsigned int data_len, - unsigned int pf, + u_int8_t pf, unsigned int hooknum, const struct net_device *indev, const struct net_device *outdev, @@ -534,7 +534,7 @@ static struct nf_loginfo default_loginfo = { /* log handler for internal netfilter logging api */ static void -nfulnl_log_packet(unsigned int pf, +nfulnl_log_packet(u_int8_t pf, unsigned int hooknum, const struct sk_buff *skb, const struct net_device *in, diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 5d75cd86ebb..89837a4eef7 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -30,7 +30,7 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); -MODULE_DESCRIPTION("[ip,ip6,arp]_tables backend module"); +MODULE_DESCRIPTION("{ip,ip6,arp,eb}_tables backend module"); #define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1)) @@ -58,17 +58,20 @@ static struct xt_af *xt; #define duprintf(format, args...) #endif -static const char *const xt_prefix[NPROTO] = { - [AF_INET] = "ip", - [AF_INET6] = "ip6", - [NF_ARP] = "arp", +static const char *const xt_prefix[NFPROTO_NUMPROTO] = { + [NFPROTO_UNSPEC] = "x", + [NFPROTO_IPV4] = "ip", + [NFPROTO_ARP] = "arp", + [NFPROTO_BRIDGE] = "eb", + [NFPROTO_IPV6] = "ip6", }; /* Registration hooks for targets. */ int xt_register_target(struct xt_target *target) { - int ret, af = target->family; + u_int8_t af = target->family; + int ret; ret = mutex_lock_interruptible(&xt[af].mutex); if (ret != 0) @@ -82,7 +85,7 @@ EXPORT_SYMBOL(xt_register_target); void xt_unregister_target(struct xt_target *target) { - int af = target->family; + u_int8_t af = target->family; mutex_lock(&xt[af].mutex); list_del(&target->list); @@ -123,7 +126,8 @@ EXPORT_SYMBOL(xt_unregister_targets); int xt_register_match(struct xt_match *match) { - int ret, af = match->family; + u_int8_t af = match->family; + int ret; ret = mutex_lock_interruptible(&xt[af].mutex); if (ret != 0) @@ -139,7 +143,7 @@ EXPORT_SYMBOL(xt_register_match); void xt_unregister_match(struct xt_match *match) { - int af = match->family; + u_int8_t af = match->family; mutex_lock(&xt[af].mutex); list_del(&match->list); @@ -185,7 +189,7 @@ EXPORT_SYMBOL(xt_unregister_matches); */ /* 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 *xt_find_match(u8 af, const char *name, u8 revision) { struct xt_match *m; int err = 0; @@ -205,12 +209,17 @@ struct xt_match *xt_find_match(int af, const char *name, u8 revision) } } mutex_unlock(&xt[af].mutex); + + if (af != NFPROTO_UNSPEC) + /* Try searching again in the family-independent list */ + return xt_find_match(NFPROTO_UNSPEC, name, revision); + 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 *xt_find_target(u8 af, const char *name, u8 revision) { struct xt_target *t; int err = 0; @@ -230,11 +239,16 @@ struct xt_target *xt_find_target(int af, const char *name, u8 revision) } } mutex_unlock(&xt[af].mutex); + + if (af != NFPROTO_UNSPEC) + /* Try searching again in the family-independent list */ + return xt_find_target(NFPROTO_UNSPEC, name, revision); + return ERR_PTR(err); } EXPORT_SYMBOL(xt_find_target); -struct xt_target *xt_request_find_target(int af, const char *name, u8 revision) +struct xt_target *xt_request_find_target(u8 af, const char *name, u8 revision) { struct xt_target *target; @@ -246,7 +260,7 @@ struct xt_target *xt_request_find_target(int af, const char *name, u8 revision) } EXPORT_SYMBOL_GPL(xt_request_find_target); -static int match_revfn(int af, const char *name, u8 revision, int *bestp) +static int match_revfn(u8 af, const char *name, u8 revision, int *bestp) { const struct xt_match *m; int have_rev = 0; @@ -262,7 +276,7 @@ static int match_revfn(int af, const char *name, u8 revision, int *bestp) return have_rev; } -static int target_revfn(int af, const char *name, u8 revision, int *bestp) +static int target_revfn(u8 af, const char *name, u8 revision, int *bestp) { const struct xt_target *t; int have_rev = 0; @@ -279,7 +293,7 @@ static int target_revfn(int af, const char *name, u8 revision, int *bestp) } /* Returns true or false (if no such extension at all) */ -int xt_find_revision(int af, const char *name, u8 revision, int target, +int xt_find_revision(u8 af, const char *name, u8 revision, int target, int *err) { int have_rev, best = -1; @@ -307,37 +321,47 @@ int xt_find_revision(int af, const char *name, u8 revision, int target, } EXPORT_SYMBOL_GPL(xt_find_revision); -int xt_check_match(const struct xt_match *match, unsigned short family, - unsigned int size, const char *table, unsigned int hook_mask, - unsigned short proto, int inv_proto) +int xt_check_match(struct xt_mtchk_param *par, + unsigned int size, u_int8_t proto, bool inv_proto) { - if (XT_ALIGN(match->matchsize) != size) { + if (XT_ALIGN(par->match->matchsize) != size && + par->match->matchsize != -1) { + /* + * ebt_among is exempt from centralized matchsize checking + * because it uses a dynamic-size data set. + */ printk("%s_tables: %s match: invalid size %Zu != %u\n", - xt_prefix[family], match->name, - XT_ALIGN(match->matchsize), size); + xt_prefix[par->family], par->match->name, + XT_ALIGN(par->match->matchsize), size); return -EINVAL; } - if (match->table && strcmp(match->table, table)) { + if (par->match->table != NULL && + strcmp(par->match->table, par->table) != 0) { printk("%s_tables: %s match: only valid in %s table, not %s\n", - xt_prefix[family], match->name, match->table, table); + xt_prefix[par->family], par->match->name, + par->match->table, par->table); return -EINVAL; } - if (match->hooks && (hook_mask & ~match->hooks) != 0) { - printk("%s_tables: %s match: bad hook_mask %u/%u\n", - xt_prefix[family], match->name, hook_mask, match->hooks); + if (par->match->hooks && (par->hook_mask & ~par->match->hooks) != 0) { + printk("%s_tables: %s match: bad hook_mask %#x/%#x\n", + xt_prefix[par->family], par->match->name, + par->hook_mask, par->match->hooks); return -EINVAL; } - if (match->proto && (match->proto != proto || inv_proto)) { + if (par->match->proto && (par->match->proto != proto || inv_proto)) { printk("%s_tables: %s match: only valid for protocol %u\n", - xt_prefix[family], match->name, match->proto); + xt_prefix[par->family], par->match->name, + par->match->proto); return -EINVAL; } + if (par->match->checkentry != NULL && !par->match->checkentry(par)) + return -EINVAL; return 0; } EXPORT_SYMBOL_GPL(xt_check_match); #ifdef CONFIG_COMPAT -int xt_compat_add_offset(int af, unsigned int offset, short delta) +int xt_compat_add_offset(u_int8_t af, unsigned int offset, short delta) { struct compat_delta *tmp; @@ -359,7 +383,7 @@ int xt_compat_add_offset(int af, unsigned int offset, short delta) } EXPORT_SYMBOL_GPL(xt_compat_add_offset); -void xt_compat_flush_offsets(int af) +void xt_compat_flush_offsets(u_int8_t af) { struct compat_delta *tmp, *next; @@ -373,7 +397,7 @@ void xt_compat_flush_offsets(int af) } EXPORT_SYMBOL_GPL(xt_compat_flush_offsets); -short xt_compat_calc_jump(int af, unsigned int offset) +short xt_compat_calc_jump(u_int8_t af, unsigned int offset) { struct compat_delta *tmp; short delta; @@ -448,32 +472,36 @@ int xt_compat_match_to_user(struct xt_entry_match *m, void __user **dstptr, EXPORT_SYMBOL_GPL(xt_compat_match_to_user); #endif /* CONFIG_COMPAT */ -int xt_check_target(const struct xt_target *target, unsigned short family, - unsigned int size, const char *table, unsigned int hook_mask, - unsigned short proto, int inv_proto) +int xt_check_target(struct xt_tgchk_param *par, + unsigned int size, u_int8_t proto, bool inv_proto) { - if (XT_ALIGN(target->targetsize) != size) { + if (XT_ALIGN(par->target->targetsize) != size) { printk("%s_tables: %s target: invalid size %Zu != %u\n", - xt_prefix[family], target->name, - XT_ALIGN(target->targetsize), size); + xt_prefix[par->family], par->target->name, + XT_ALIGN(par->target->targetsize), size); return -EINVAL; } - if (target->table && strcmp(target->table, table)) { + if (par->target->table != NULL && + strcmp(par->target->table, par->table) != 0) { printk("%s_tables: %s target: only valid in %s table, not %s\n", - xt_prefix[family], target->name, target->table, table); + xt_prefix[par->family], par->target->name, + par->target->table, par->table); return -EINVAL; } - if (target->hooks && (hook_mask & ~target->hooks) != 0) { - printk("%s_tables: %s target: bad hook_mask %u/%u\n", - xt_prefix[family], target->name, hook_mask, - target->hooks); + if (par->target->hooks && (par->hook_mask & ~par->target->hooks) != 0) { + printk("%s_tables: %s target: bad hook_mask %#x/%#x\n", + xt_prefix[par->family], par->target->name, + par->hook_mask, par->target->hooks); return -EINVAL; } - if (target->proto && (target->proto != proto || inv_proto)) { + if (par->target->proto && (par->target->proto != proto || inv_proto)) { printk("%s_tables: %s target: only valid for protocol %u\n", - xt_prefix[family], target->name, target->proto); + xt_prefix[par->family], par->target->name, + par->target->proto); return -EINVAL; } + if (par->target->checkentry != NULL && !par->target->checkentry(par)) + return -EINVAL; return 0; } EXPORT_SYMBOL_GPL(xt_check_target); @@ -590,7 +618,8 @@ void xt_free_table_info(struct xt_table_info *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(struct net *net, int af, const char *name) +struct xt_table *xt_find_table_lock(struct net *net, u_int8_t af, + const char *name) { struct xt_table *t; @@ -612,13 +641,13 @@ void xt_table_unlock(struct xt_table *table) EXPORT_SYMBOL_GPL(xt_table_unlock); #ifdef CONFIG_COMPAT -void xt_compat_lock(int af) +void xt_compat_lock(u_int8_t af) { mutex_lock(&xt[af].compat_mutex); } EXPORT_SYMBOL_GPL(xt_compat_lock); -void xt_compat_unlock(int af) +void xt_compat_unlock(u_int8_t af) { mutex_unlock(&xt[af].compat_mutex); } @@ -722,13 +751,13 @@ EXPORT_SYMBOL_GPL(xt_unregister_table); #ifdef CONFIG_PROC_FS struct xt_names_priv { struct seq_net_private p; - int af; + u_int8_t af; }; static void *xt_table_seq_start(struct seq_file *seq, loff_t *pos) { struct xt_names_priv *priv = seq->private; struct net *net = seq_file_net(seq); - int af = priv->af; + u_int8_t af = priv->af; mutex_lock(&xt[af].mutex); return seq_list_start(&net->xt.tables[af], *pos); @@ -738,7 +767,7 @@ static void *xt_table_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct xt_names_priv *priv = seq->private; struct net *net = seq_file_net(seq); - int af = priv->af; + u_int8_t af = priv->af; return seq_list_next(v, &net->xt.tables[af], pos); } @@ -746,7 +775,7 @@ static void *xt_table_seq_next(struct seq_file *seq, void *v, loff_t *pos) static void xt_table_seq_stop(struct seq_file *seq, void *v) { struct xt_names_priv *priv = seq->private; - int af = priv->af; + u_int8_t af = priv->af; mutex_unlock(&xt[af].mutex); } @@ -922,14 +951,14 @@ static const struct file_operations xt_target_ops = { #endif /* CONFIG_PROC_FS */ -int xt_proto_init(struct net *net, int af) +int xt_proto_init(struct net *net, u_int8_t af) { #ifdef CONFIG_PROC_FS char buf[XT_FUNCTION_MAXNAMELEN]; struct proc_dir_entry *proc; #endif - if (af >= NPROTO) + if (af >= ARRAY_SIZE(xt_prefix)) return -EINVAL; @@ -974,7 +1003,7 @@ out: } EXPORT_SYMBOL_GPL(xt_proto_init); -void xt_proto_fini(struct net *net, int af) +void xt_proto_fini(struct net *net, u_int8_t af) { #ifdef CONFIG_PROC_FS char buf[XT_FUNCTION_MAXNAMELEN]; @@ -998,7 +1027,7 @@ static int __net_init xt_net_init(struct net *net) { int i; - for (i = 0; i < NPROTO; i++) + for (i = 0; i < NFPROTO_NUMPROTO; i++) INIT_LIST_HEAD(&net->xt.tables[i]); return 0; } @@ -1011,11 +1040,11 @@ static int __init xt_init(void) { int i, rv; - xt = kmalloc(sizeof(struct xt_af) * NPROTO, GFP_KERNEL); + xt = kmalloc(sizeof(struct xt_af) * NFPROTO_NUMPROTO, GFP_KERNEL); if (!xt) return -ENOMEM; - for (i = 0; i < NPROTO; i++) { + for (i = 0; i < NFPROTO_NUMPROTO; i++) { mutex_init(&xt[i].mutex); #ifdef CONFIG_COMPAT mutex_init(&xt[i].compat_mutex); diff --git a/net/netfilter/xt_CLASSIFY.c b/net/netfilter/xt_CLASSIFY.c index 77a52bf8322..011bc80dd2a 100644 --- a/net/netfilter/xt_CLASSIFY.c +++ b/net/netfilter/xt_CLASSIFY.c @@ -27,50 +27,34 @@ MODULE_ALIAS("ipt_CLASSIFY"); MODULE_ALIAS("ip6t_CLASSIFY"); static unsigned int -classify_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +classify_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_classify_target_info *clinfo = targinfo; + const struct xt_classify_target_info *clinfo = par->targinfo; skb->priority = clinfo->priority; return XT_CONTINUE; } -static struct xt_target classify_tg_reg[] __read_mostly = { - { - .family = AF_INET, - .name = "CLASSIFY", - .target = classify_tg, - .targetsize = sizeof(struct xt_classify_target_info), - .table = "mangle", - .hooks = (1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_FORWARD) | - (1 << NF_INET_POST_ROUTING), - .me = THIS_MODULE, - }, - { - .name = "CLASSIFY", - .family = AF_INET6, - .target = classify_tg, - .targetsize = sizeof(struct xt_classify_target_info), - .table = "mangle", - .hooks = (1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_FORWARD) | - (1 << NF_INET_POST_ROUTING), - .me = THIS_MODULE, - }, +static struct xt_target classify_tg_reg __read_mostly = { + .name = "CLASSIFY", + .revision = 0, + .family = NFPROTO_UNSPEC, + .table = "mangle", + .hooks = (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_FORWARD) | + (1 << NF_INET_POST_ROUTING), + .target = classify_tg, + .targetsize = sizeof(struct xt_classify_target_info), + .me = THIS_MODULE, }; static int __init classify_tg_init(void) { - return xt_register_targets(classify_tg_reg, - ARRAY_SIZE(classify_tg_reg)); + return xt_register_target(&classify_tg_reg); } static void __exit classify_tg_exit(void) { - xt_unregister_targets(classify_tg_reg, ARRAY_SIZE(classify_tg_reg)); + xt_unregister_target(&classify_tg_reg); } module_init(classify_tg_init); diff --git a/net/netfilter/xt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c index 5fecfb4794b..d6e5ab46327 100644 --- a/net/netfilter/xt_CONNMARK.c +++ b/net/netfilter/xt_CONNMARK.c @@ -36,11 +36,9 @@ MODULE_ALIAS("ip6t_CONNMARK"); #include <net/netfilter/nf_conntrack_ecache.h> static unsigned int -connmark_tg_v0(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +connmark_tg_v0(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_connmark_target_info *markinfo = targinfo; + const struct xt_connmark_target_info *markinfo = par->targinfo; struct nf_conn *ct; enum ip_conntrack_info ctinfo; u_int32_t diff; @@ -54,7 +52,7 @@ connmark_tg_v0(struct sk_buff *skb, const struct net_device *in, newmark = (ct->mark & ~markinfo->mask) | markinfo->mark; if (newmark != ct->mark) { ct->mark = newmark; - nf_conntrack_event_cache(IPCT_MARK, skb); + nf_conntrack_event_cache(IPCT_MARK, ct); } break; case XT_CONNMARK_SAVE: @@ -62,7 +60,7 @@ connmark_tg_v0(struct sk_buff *skb, const struct net_device *in, (skb->mark & markinfo->mask); if (ct->mark != newmark) { ct->mark = newmark; - nf_conntrack_event_cache(IPCT_MARK, skb); + nf_conntrack_event_cache(IPCT_MARK, ct); } break; case XT_CONNMARK_RESTORE: @@ -77,11 +75,9 @@ connmark_tg_v0(struct sk_buff *skb, const struct net_device *in, } static unsigned int -connmark_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +connmark_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_connmark_tginfo1 *info = targinfo; + const struct xt_connmark_tginfo1 *info = par->targinfo; enum ip_conntrack_info ctinfo; struct nf_conn *ct; u_int32_t newmark; @@ -95,7 +91,7 @@ connmark_tg(struct sk_buff *skb, const struct net_device *in, newmark = (ct->mark & ~info->ctmask) ^ info->ctmark; if (ct->mark != newmark) { ct->mark = newmark; - nf_conntrack_event_cache(IPCT_MARK, skb); + nf_conntrack_event_cache(IPCT_MARK, ct); } break; case XT_CONNMARK_SAVE: @@ -103,7 +99,7 @@ connmark_tg(struct sk_buff *skb, const struct net_device *in, (skb->mark & info->nfmask); if (ct->mark != newmark) { ct->mark = newmark; - nf_conntrack_event_cache(IPCT_MARK, skb); + nf_conntrack_event_cache(IPCT_MARK, ct); } break; case XT_CONNMARK_RESTORE: @@ -116,18 +112,15 @@ connmark_tg(struct sk_buff *skb, const struct net_device *in, return XT_CONTINUE; } -static bool -connmark_tg_check_v0(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool connmark_tg_check_v0(const struct xt_tgchk_param *par) { - const struct xt_connmark_target_info *matchinfo = targinfo; + const struct xt_connmark_target_info *matchinfo = par->targinfo; if (matchinfo->mode == XT_CONNMARK_RESTORE) { - if (strcmp(tablename, "mangle") != 0) { + if (strcmp(par->table, "mangle") != 0) { printk(KERN_WARNING "CONNMARK: restore can only be " "called from \"mangle\" table, not \"%s\"\n", - tablename); + par->table); return false; } } @@ -135,31 +128,27 @@ connmark_tg_check_v0(const char *tablename, const void *entry, printk(KERN_WARNING "CONNMARK: Only supports 32bit mark\n"); return false; } - if (nf_ct_l3proto_try_module_get(target->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", target->family); + "proto=%u\n", par->family); return false; } return true; } -static bool -connmark_tg_check(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool connmark_tg_check(const struct xt_tgchk_param *par) { - if (nf_ct_l3proto_try_module_get(target->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "cannot load conntrack support for " - "proto=%u\n", target->family); + "proto=%u\n", par->family); return false; } return true; } -static void -connmark_tg_destroy(const struct xt_target *target, void *targinfo) +static void connmark_tg_destroy(const struct xt_tgdtor_param *par) { - nf_ct_l3proto_module_put(target->family); + nf_ct_l3proto_module_put(par->family); } #ifdef CONFIG_COMPAT @@ -197,7 +186,7 @@ static struct xt_target connmark_tg_reg[] __read_mostly = { { .name = "CONNMARK", .revision = 0, - .family = AF_INET, + .family = NFPROTO_UNSPEC, .checkentry = connmark_tg_check_v0, .destroy = connmark_tg_destroy, .target = connmark_tg_v0, @@ -210,34 +199,9 @@ static struct xt_target connmark_tg_reg[] __read_mostly = { .me = THIS_MODULE }, { - .name = "CONNMARK", - .revision = 0, - .family = AF_INET6, - .checkentry = connmark_tg_check_v0, - .destroy = connmark_tg_destroy, - .target = connmark_tg_v0, - .targetsize = sizeof(struct xt_connmark_target_info), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_connmark_target_info), - .compat_from_user = connmark_tg_compat_from_user_v0, - .compat_to_user = connmark_tg_compat_to_user_v0, -#endif - .me = THIS_MODULE - }, - { - .name = "CONNMARK", - .revision = 1, - .family = AF_INET, - .checkentry = connmark_tg_check, - .target = connmark_tg, - .targetsize = sizeof(struct xt_connmark_tginfo1), - .destroy = connmark_tg_destroy, - .me = THIS_MODULE, - }, - { .name = "CONNMARK", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .checkentry = connmark_tg_check, .target = connmark_tg, .targetsize = sizeof(struct xt_connmark_tginfo1), diff --git a/net/netfilter/xt_CONNSECMARK.c b/net/netfilter/xt_CONNSECMARK.c index 76ca1f2421e..b54c3756fdc 100644 --- a/net/netfilter/xt_CONNSECMARK.c +++ b/net/netfilter/xt_CONNSECMARK.c @@ -43,7 +43,7 @@ static void secmark_save(const struct sk_buff *skb) ct = nf_ct_get(skb, &ctinfo); if (ct && !ct->secmark) { ct->secmark = skb->secmark; - nf_conntrack_event_cache(IPCT_SECMARK, skb); + nf_conntrack_event_cache(IPCT_SECMARK, ct); } } } @@ -65,11 +65,9 @@ static void secmark_restore(struct sk_buff *skb) } static unsigned int -connsecmark_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +connsecmark_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_connsecmark_target_info *info = targinfo; + const struct xt_connsecmark_target_info *info = par->targinfo; switch (info->mode) { case CONNSECMARK_SAVE: @@ -87,16 +85,14 @@ connsecmark_tg(struct sk_buff *skb, const struct net_device *in, return XT_CONTINUE; } -static bool -connsecmark_tg_check(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool connsecmark_tg_check(const struct xt_tgchk_param *par) { - const struct xt_connsecmark_target_info *info = targinfo; + const struct xt_connsecmark_target_info *info = par->targinfo; - if (strcmp(tablename, "mangle") && strcmp(tablename, "security")) { + if (strcmp(par->table, "mangle") != 0 && + strcmp(par->table, "security") != 0) { printk(KERN_INFO PFX "target only valid in the \'mangle\' " - "or \'security\' tables, not \'%s\'.\n", tablename); + "or \'security\' tables, not \'%s\'.\n", par->table); return false; } @@ -110,51 +106,38 @@ connsecmark_tg_check(const char *tablename, const void *entry, return false; } - if (nf_ct_l3proto_try_module_get(target->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", target->family); + "proto=%u\n", par->family); return false; } return true; } -static void -connsecmark_tg_destroy(const struct xt_target *target, void *targinfo) +static void connsecmark_tg_destroy(const struct xt_tgdtor_param *par) { - nf_ct_l3proto_module_put(target->family); + nf_ct_l3proto_module_put(par->family); } -static struct xt_target connsecmark_tg_reg[] __read_mostly = { - { - .name = "CONNSECMARK", - .family = AF_INET, - .checkentry = connsecmark_tg_check, - .destroy = connsecmark_tg_destroy, - .target = connsecmark_tg, - .targetsize = sizeof(struct xt_connsecmark_target_info), - .me = THIS_MODULE, - }, - { - .name = "CONNSECMARK", - .family = AF_INET6, - .checkentry = connsecmark_tg_check, - .destroy = connsecmark_tg_destroy, - .target = connsecmark_tg, - .targetsize = sizeof(struct xt_connsecmark_target_info), - .me = THIS_MODULE, - }, +static struct xt_target connsecmark_tg_reg __read_mostly = { + .name = "CONNSECMARK", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = connsecmark_tg_check, + .destroy = connsecmark_tg_destroy, + .target = connsecmark_tg, + .targetsize = sizeof(struct xt_connsecmark_target_info), + .me = THIS_MODULE, }; static int __init connsecmark_tg_init(void) { - return xt_register_targets(connsecmark_tg_reg, - ARRAY_SIZE(connsecmark_tg_reg)); + return xt_register_target(&connsecmark_tg_reg); } static void __exit connsecmark_tg_exit(void) { - xt_unregister_targets(connsecmark_tg_reg, - ARRAY_SIZE(connsecmark_tg_reg)); + xt_unregister_target(&connsecmark_tg_reg); } module_init(connsecmark_tg_init); diff --git a/net/netfilter/xt_DSCP.c b/net/netfilter/xt_DSCP.c index 97efd74c04f..6a347e768f8 100644 --- a/net/netfilter/xt_DSCP.c +++ b/net/netfilter/xt_DSCP.c @@ -29,11 +29,9 @@ MODULE_ALIAS("ipt_TOS"); MODULE_ALIAS("ip6t_TOS"); static unsigned int -dscp_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +dscp_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_DSCP_info *dinfo = targinfo; + const struct xt_DSCP_info *dinfo = par->targinfo; u_int8_t dscp = ipv4_get_dsfield(ip_hdr(skb)) >> XT_DSCP_SHIFT; if (dscp != dinfo->dscp) { @@ -48,11 +46,9 @@ dscp_tg(struct sk_buff *skb, const struct net_device *in, } static unsigned int -dscp_tg6(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +dscp_tg6(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_DSCP_info *dinfo = targinfo; + const struct xt_DSCP_info *dinfo = par->targinfo; u_int8_t dscp = ipv6_get_dsfield(ipv6_hdr(skb)) >> XT_DSCP_SHIFT; if (dscp != dinfo->dscp) { @@ -65,26 +61,21 @@ dscp_tg6(struct sk_buff *skb, const struct net_device *in, return XT_CONTINUE; } -static bool -dscp_tg_check(const char *tablename, const void *e_void, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool dscp_tg_check(const struct xt_tgchk_param *par) { - const u_int8_t dscp = ((struct xt_DSCP_info *)targinfo)->dscp; + const struct xt_DSCP_info *info = par->targinfo; - if (dscp > XT_DSCP_MAX) { - printk(KERN_WARNING "DSCP: dscp %x out of range\n", dscp); + if (info->dscp > XT_DSCP_MAX) { + printk(KERN_WARNING "DSCP: dscp %x out of range\n", info->dscp); return false; } return true; } static unsigned int -tos_tg_v0(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tos_tg_v0(struct sk_buff *skb, const struct xt_target_param *par) { - const struct ipt_tos_target_info *info = targinfo; + const struct ipt_tos_target_info *info = par->targinfo; struct iphdr *iph = ip_hdr(skb); u_int8_t oldtos; @@ -101,12 +92,10 @@ tos_tg_v0(struct sk_buff *skb, const struct net_device *in, return XT_CONTINUE; } -static bool -tos_tg_check_v0(const char *tablename, const void *e_void, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool tos_tg_check_v0(const struct xt_tgchk_param *par) { - const u_int8_t tos = ((struct ipt_tos_target_info *)targinfo)->tos; + const struct ipt_tos_target_info *info = par->targinfo; + const uint8_t tos = info->tos; if (tos != IPTOS_LOWDELAY && tos != IPTOS_THROUGHPUT && tos != IPTOS_RELIABILITY && tos != IPTOS_MINCOST && @@ -119,11 +108,9 @@ tos_tg_check_v0(const char *tablename, const void *e_void, } static unsigned int -tos_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tos_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_tos_target_info *info = targinfo; + const struct xt_tos_target_info *info = par->targinfo; struct iphdr *iph = ip_hdr(skb); u_int8_t orig, nv; @@ -141,11 +128,9 @@ tos_tg(struct sk_buff *skb, const struct net_device *in, } static unsigned int -tos_tg6(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tos_tg6(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_tos_target_info *info = targinfo; + const struct xt_tos_target_info *info = par->targinfo; struct ipv6hdr *iph = ipv6_hdr(skb); u_int8_t orig, nv; @@ -165,7 +150,7 @@ tos_tg6(struct sk_buff *skb, const struct net_device *in, static struct xt_target dscp_tg_reg[] __read_mostly = { { .name = "DSCP", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = dscp_tg_check, .target = dscp_tg, .targetsize = sizeof(struct xt_DSCP_info), @@ -174,7 +159,7 @@ static struct xt_target dscp_tg_reg[] __read_mostly = { }, { .name = "DSCP", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = dscp_tg_check, .target = dscp_tg6, .targetsize = sizeof(struct xt_DSCP_info), @@ -184,7 +169,7 @@ static struct xt_target dscp_tg_reg[] __read_mostly = { { .name = "TOS", .revision = 0, - .family = AF_INET, + .family = NFPROTO_IPV4, .table = "mangle", .target = tos_tg_v0, .targetsize = sizeof(struct ipt_tos_target_info), @@ -194,7 +179,7 @@ static struct xt_target dscp_tg_reg[] __read_mostly = { { .name = "TOS", .revision = 1, - .family = AF_INET, + .family = NFPROTO_IPV4, .table = "mangle", .target = tos_tg, .targetsize = sizeof(struct xt_tos_target_info), @@ -203,7 +188,7 @@ static struct xt_target dscp_tg_reg[] __read_mostly = { { .name = "TOS", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_IPV6, .table = "mangle", .target = tos_tg6, .targetsize = sizeof(struct xt_tos_target_info), diff --git a/net/netfilter/xt_MARK.c b/net/netfilter/xt_MARK.c index f9ce20b5898..67574bcfb8a 100644 --- a/net/netfilter/xt_MARK.c +++ b/net/netfilter/xt_MARK.c @@ -25,22 +25,18 @@ MODULE_ALIAS("ipt_MARK"); MODULE_ALIAS("ip6t_MARK"); static unsigned int -mark_tg_v0(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +mark_tg_v0(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_mark_target_info *markinfo = targinfo; + const struct xt_mark_target_info *markinfo = par->targinfo; skb->mark = markinfo->mark; return XT_CONTINUE; } static unsigned int -mark_tg_v1(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +mark_tg_v1(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_mark_target_info_v1 *markinfo = targinfo; + const struct xt_mark_target_info_v1 *markinfo = par->targinfo; int mark = 0; switch (markinfo->mode) { @@ -62,22 +58,17 @@ mark_tg_v1(struct sk_buff *skb, const struct net_device *in, } static unsigned int -mark_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +mark_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_mark_tginfo2 *info = targinfo; + const struct xt_mark_tginfo2 *info = par->targinfo; skb->mark = (skb->mark & ~info->mask) ^ info->mark; return XT_CONTINUE; } -static bool -mark_tg_check_v0(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool mark_tg_check_v0(const struct xt_tgchk_param *par) { - const struct xt_mark_target_info *markinfo = targinfo; + const struct xt_mark_target_info *markinfo = par->targinfo; if (markinfo->mark > 0xffffffff) { printk(KERN_WARNING "MARK: Only supports 32bit wide mark\n"); @@ -86,12 +77,9 @@ mark_tg_check_v0(const char *tablename, const void *entry, return true; } -static bool -mark_tg_check_v1(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool mark_tg_check_v1(const struct xt_tgchk_param *par) { - const struct xt_mark_target_info_v1 *markinfo = targinfo; + const struct xt_mark_target_info_v1 *markinfo = par->targinfo; if (markinfo->mode != XT_MARK_SET && markinfo->mode != XT_MARK_AND @@ -161,7 +149,7 @@ static int mark_tg_compat_to_user_v1(void __user *dst, void *src) static struct xt_target mark_tg_reg[] __read_mostly = { { .name = "MARK", - .family = AF_INET, + .family = NFPROTO_UNSPEC, .revision = 0, .checkentry = mark_tg_check_v0, .target = mark_tg_v0, @@ -176,7 +164,7 @@ static struct xt_target mark_tg_reg[] __read_mostly = { }, { .name = "MARK", - .family = AF_INET, + .family = NFPROTO_UNSPEC, .revision = 1, .checkentry = mark_tg_check_v1, .target = mark_tg_v1, @@ -190,47 +178,9 @@ static struct xt_target mark_tg_reg[] __read_mostly = { .me = THIS_MODULE, }, { - .name = "MARK", - .family = AF_INET6, - .revision = 0, - .checkentry = mark_tg_check_v0, - .target = mark_tg_v0, - .targetsize = sizeof(struct xt_mark_target_info), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_mark_target_info), - .compat_from_user = mark_tg_compat_from_user_v0, - .compat_to_user = mark_tg_compat_to_user_v0, -#endif - .table = "mangle", - .me = THIS_MODULE, - }, - { - .name = "MARK", - .family = AF_INET6, - .revision = 1, - .checkentry = mark_tg_check_v1, - .target = mark_tg_v1, - .targetsize = sizeof(struct xt_mark_target_info_v1), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_mark_target_info_v1), - .compat_from_user = mark_tg_compat_from_user_v1, - .compat_to_user = mark_tg_compat_to_user_v1, -#endif - .table = "mangle", - .me = THIS_MODULE, - }, - { - .name = "MARK", - .revision = 2, - .family = AF_INET, - .target = mark_tg, - .targetsize = sizeof(struct xt_mark_tginfo2), - .me = THIS_MODULE, - }, - { .name = "MARK", .revision = 2, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .target = mark_tg, .targetsize = sizeof(struct xt_mark_tginfo2), .me = THIS_MODULE, diff --git a/net/netfilter/xt_NFLOG.c b/net/netfilter/xt_NFLOG.c index 19ae8efae65..50e3a52d3b3 100644 --- a/net/netfilter/xt_NFLOG.c +++ b/net/netfilter/xt_NFLOG.c @@ -21,11 +21,9 @@ MODULE_ALIAS("ipt_NFLOG"); MODULE_ALIAS("ip6t_NFLOG"); static unsigned int -nflog_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +nflog_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_nflog_info *info = targinfo; + const struct xt_nflog_info *info = par->targinfo; struct nf_loginfo li; li.type = NF_LOG_TYPE_ULOG; @@ -33,17 +31,14 @@ nflog_tg(struct sk_buff *skb, const struct net_device *in, li.u.ulog.group = info->group; li.u.ulog.qthreshold = info->threshold; - nf_log_packet(target->family, hooknum, skb, in, out, &li, - "%s", info->prefix); + nf_log_packet(par->family, par->hooknum, skb, par->in, + par->out, &li, "%s", info->prefix); return XT_CONTINUE; } -static bool -nflog_tg_check(const char *tablename, const void *entry, - const struct xt_target *target, void *targetinfo, - unsigned int hookmask) +static bool nflog_tg_check(const struct xt_tgchk_param *par) { - const struct xt_nflog_info *info = targetinfo; + const struct xt_nflog_info *info = par->targinfo; if (info->flags & ~XT_NFLOG_MASK) return false; @@ -52,33 +47,24 @@ nflog_tg_check(const char *tablename, const void *entry, return true; } -static struct xt_target nflog_tg_reg[] __read_mostly = { - { - .name = "NFLOG", - .family = AF_INET, - .checkentry = nflog_tg_check, - .target = nflog_tg, - .targetsize = sizeof(struct xt_nflog_info), - .me = THIS_MODULE, - }, - { - .name = "NFLOG", - .family = AF_INET6, - .checkentry = nflog_tg_check, - .target = nflog_tg, - .targetsize = sizeof(struct xt_nflog_info), - .me = THIS_MODULE, - }, +static struct xt_target nflog_tg_reg __read_mostly = { + .name = "NFLOG", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = nflog_tg_check, + .target = nflog_tg, + .targetsize = sizeof(struct xt_nflog_info), + .me = THIS_MODULE, }; static int __init nflog_tg_init(void) { - return xt_register_targets(nflog_tg_reg, ARRAY_SIZE(nflog_tg_reg)); + return xt_register_target(&nflog_tg_reg); } static void __exit nflog_tg_exit(void) { - xt_unregister_targets(nflog_tg_reg, ARRAY_SIZE(nflog_tg_reg)); + xt_unregister_target(&nflog_tg_reg); } module_init(nflog_tg_init); diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index beb24d19a56..2cc1fff4930 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -24,11 +24,9 @@ MODULE_ALIAS("ip6t_NFQUEUE"); MODULE_ALIAS("arpt_NFQUEUE"); static unsigned int -nfqueue_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +nfqueue_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_NFQ_info *tinfo = targinfo; + const struct xt_NFQ_info *tinfo = par->targinfo; return NF_QUEUE_NR(tinfo->queuenum); } @@ -36,14 +34,14 @@ nfqueue_tg(struct sk_buff *skb, const struct net_device *in, static struct xt_target nfqueue_tg_reg[] __read_mostly = { { .name = "NFQUEUE", - .family = AF_INET, + .family = NFPROTO_IPV4, .target = nfqueue_tg, .targetsize = sizeof(struct xt_NFQ_info), .me = THIS_MODULE, }, { .name = "NFQUEUE", - .family = AF_INET6, + .family = NFPROTO_IPV6, .target = nfqueue_tg, .targetsize = sizeof(struct xt_NFQ_info), .me = THIS_MODULE, diff --git a/net/netfilter/xt_NOTRACK.c b/net/netfilter/xt_NOTRACK.c index 6c9de611eb8..e7a0a54fd4e 100644 --- a/net/netfilter/xt_NOTRACK.c +++ b/net/netfilter/xt_NOTRACK.c @@ -13,9 +13,7 @@ MODULE_ALIAS("ipt_NOTRACK"); MODULE_ALIAS("ip6t_NOTRACK"); static unsigned int -notrack_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +notrack_tg(struct sk_buff *skb, const struct xt_target_param *par) { /* Previously seen (loopback)? Ignore. */ if (skb->nfct != NULL) @@ -32,31 +30,23 @@ notrack_tg(struct sk_buff *skb, const struct net_device *in, return XT_CONTINUE; } -static struct xt_target notrack_tg_reg[] __read_mostly = { - { - .name = "NOTRACK", - .family = AF_INET, - .target = notrack_tg, - .table = "raw", - .me = THIS_MODULE, - }, - { - .name = "NOTRACK", - .family = AF_INET6, - .target = notrack_tg, - .table = "raw", - .me = THIS_MODULE, - }, +static struct xt_target notrack_tg_reg __read_mostly = { + .name = "NOTRACK", + .revision = 0, + .family = NFPROTO_UNSPEC, + .target = notrack_tg, + .table = "raw", + .me = THIS_MODULE, }; static int __init notrack_tg_init(void) { - return xt_register_targets(notrack_tg_reg, ARRAY_SIZE(notrack_tg_reg)); + return xt_register_target(¬rack_tg_reg); } static void __exit notrack_tg_exit(void) { - xt_unregister_targets(notrack_tg_reg, ARRAY_SIZE(notrack_tg_reg)); + xt_unregister_target(¬rack_tg_reg); } module_init(notrack_tg_init); diff --git a/net/netfilter/xt_RATEEST.c b/net/netfilter/xt_RATEEST.c index 64d6ad38029..43f5676b1af 100644 --- a/net/netfilter/xt_RATEEST.c +++ b/net/netfilter/xt_RATEEST.c @@ -71,14 +71,9 @@ void xt_rateest_put(struct xt_rateest *est) EXPORT_SYMBOL_GPL(xt_rateest_put); static unsigned int -xt_rateest_tg(struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - unsigned int hooknum, - const struct xt_target *target, - const void *targinfo) +xt_rateest_tg(struct sk_buff *skb, const struct xt_target_param *par) { - const struct xt_rateest_target_info *info = targinfo; + const struct xt_rateest_target_info *info = par->targinfo; struct gnet_stats_basic *stats = &info->est->bstats; spin_lock_bh(&info->est->lock); @@ -89,14 +84,9 @@ xt_rateest_tg(struct sk_buff *skb, return XT_CONTINUE; } -static bool -xt_rateest_tg_checkentry(const char *tablename, - const void *entry, - const struct xt_target *target, - void *targinfo, - unsigned int hook_mask) +static bool xt_rateest_tg_checkentry(const struct xt_tgchk_param *par) { - struct xt_rateest_target_info *info = targinfo; + struct xt_rateest_target_info *info = par->targinfo; struct xt_rateest *est; struct { struct nlattr opt; @@ -149,33 +139,22 @@ err1: return false; } -static void xt_rateest_tg_destroy(const struct xt_target *target, - void *targinfo) +static void xt_rateest_tg_destroy(const struct xt_tgdtor_param *par) { - struct xt_rateest_target_info *info = targinfo; + struct xt_rateest_target_info *info = par->targinfo; xt_rateest_put(info->est); } -static struct xt_target xt_rateest_target[] __read_mostly = { - { - .family = AF_INET, - .name = "RATEEST", - .target = xt_rateest_tg, - .checkentry = xt_rateest_tg_checkentry, - .destroy = xt_rateest_tg_destroy, - .targetsize = sizeof(struct xt_rateest_target_info), - .me = THIS_MODULE, - }, - { - .family = AF_INET6, - .name = "RATEEST", - .target = xt_rateest_tg, - .checkentry = xt_rateest_tg_checkentry, - .destroy = xt_rateest_tg_destroy, - .targetsize = sizeof(struct xt_rateest_target_info), - .me = THIS_MODULE, - }, +static struct xt_target xt_rateest_tg_reg __read_mostly = { + .name = "RATEEST", + .revision = 0, + .family = NFPROTO_UNSPEC, + .target = xt_rateest_tg, + .checkentry = xt_rateest_tg_checkentry, + .destroy = xt_rateest_tg_destroy, + .targetsize = sizeof(struct xt_rateest_target_info), + .me = THIS_MODULE, }; static int __init xt_rateest_tg_init(void) @@ -186,13 +165,12 @@ static int __init xt_rateest_tg_init(void) INIT_HLIST_HEAD(&rateest_hash[i]); get_random_bytes(&jhash_rnd, sizeof(jhash_rnd)); - return xt_register_targets(xt_rateest_target, - ARRAY_SIZE(xt_rateest_target)); + return xt_register_target(&xt_rateest_tg_reg); } static void __exit xt_rateest_tg_fini(void) { - xt_unregister_targets(xt_rateest_target, ARRAY_SIZE(xt_rateest_target)); + xt_unregister_target(&xt_rateest_tg_reg); } diff --git a/net/netfilter/xt_SECMARK.c b/net/netfilter/xt_SECMARK.c index 94f87ee7552..7a6f9e6f5df 100644 --- a/net/netfilter/xt_SECMARK.c +++ b/net/netfilter/xt_SECMARK.c @@ -29,12 +29,10 @@ MODULE_ALIAS("ip6t_SECMARK"); static u8 mode; static unsigned int -secmark_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +secmark_tg(struct sk_buff *skb, const struct xt_target_param *par) { u32 secmark = 0; - const struct xt_secmark_target_info *info = targinfo; + const struct xt_secmark_target_info *info = par->targinfo; BUG_ON(info->mode != mode); @@ -82,16 +80,14 @@ static bool checkentry_selinux(struct xt_secmark_target_info *info) return true; } -static bool -secmark_tg_check(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool secmark_tg_check(const struct xt_tgchk_param *par) { - struct xt_secmark_target_info *info = targinfo; + struct xt_secmark_target_info *info = par->targinfo; - if (strcmp(tablename, "mangle") && strcmp(tablename, "security")) { + if (strcmp(par->table, "mangle") != 0 && + strcmp(par->table, "security") != 0) { printk(KERN_INFO PFX "target only valid in the \'mangle\' " - "or \'security\' tables, not \'%s\'.\n", tablename); + "or \'security\' tables, not \'%s\'.\n", par->table); return false; } @@ -117,7 +113,7 @@ secmark_tg_check(const char *tablename, const void *entry, return true; } -static void secmark_tg_destroy(const struct xt_target *target, void *targinfo) +static void secmark_tg_destroy(const struct xt_tgdtor_param *par) { switch (mode) { case SECMARK_MODE_SEL: @@ -125,35 +121,25 @@ static void secmark_tg_destroy(const struct xt_target *target, void *targinfo) } } -static struct xt_target secmark_tg_reg[] __read_mostly = { - { - .name = "SECMARK", - .family = AF_INET, - .checkentry = secmark_tg_check, - .destroy = secmark_tg_destroy, - .target = secmark_tg, - .targetsize = sizeof(struct xt_secmark_target_info), - .me = THIS_MODULE, - }, - { - .name = "SECMARK", - .family = AF_INET6, - .checkentry = secmark_tg_check, - .destroy = secmark_tg_destroy, - .target = secmark_tg, - .targetsize = sizeof(struct xt_secmark_target_info), - .me = THIS_MODULE, - }, +static struct xt_target secmark_tg_reg __read_mostly = { + .name = "SECMARK", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = secmark_tg_check, + .destroy = secmark_tg_destroy, + .target = secmark_tg, + .targetsize = sizeof(struct xt_secmark_target_info), + .me = THIS_MODULE, }; static int __init secmark_tg_init(void) { - return xt_register_targets(secmark_tg_reg, ARRAY_SIZE(secmark_tg_reg)); + return xt_register_target(&secmark_tg_reg); } static void __exit secmark_tg_exit(void) { - xt_unregister_targets(secmark_tg_reg, ARRAY_SIZE(secmark_tg_reg)); + xt_unregister_target(&secmark_tg_reg); } module_init(secmark_tg_init); diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c index beb5094703c..4f3b1f80879 100644 --- a/net/netfilter/xt_TCPMSS.c +++ b/net/netfilter/xt_TCPMSS.c @@ -174,15 +174,13 @@ static u_int32_t tcpmss_reverse_mtu(const struct sk_buff *skb, } static unsigned int -tcpmss_tg4(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tcpmss_tg4(struct sk_buff *skb, const struct xt_target_param *par) { struct iphdr *iph = ip_hdr(skb); __be16 newlen; int ret; - ret = tcpmss_mangle_packet(skb, targinfo, + ret = tcpmss_mangle_packet(skb, par->targinfo, tcpmss_reverse_mtu(skb, PF_INET), iph->ihl * 4, sizeof(*iph) + sizeof(struct tcphdr)); @@ -199,9 +197,7 @@ tcpmss_tg4(struct sk_buff *skb, const struct net_device *in, #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) static unsigned int -tcpmss_tg6(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tcpmss_tg6(struct sk_buff *skb, const struct xt_target_param *par) { struct ipv6hdr *ipv6h = ipv6_hdr(skb); u8 nexthdr; @@ -212,7 +208,7 @@ tcpmss_tg6(struct sk_buff *skb, const struct net_device *in, tcphoff = ipv6_skip_exthdr(skb, sizeof(*ipv6h), &nexthdr); if (tcphoff < 0) return NF_DROP; - ret = tcpmss_mangle_packet(skb, targinfo, + ret = tcpmss_mangle_packet(skb, par->targinfo, tcpmss_reverse_mtu(skb, PF_INET6), tcphoff, sizeof(*ipv6h) + sizeof(struct tcphdr)); @@ -241,16 +237,13 @@ static inline bool find_syn_match(const struct xt_entry_match *m) return false; } -static bool -tcpmss_tg4_check(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool tcpmss_tg4_check(const struct xt_tgchk_param *par) { - const struct xt_tcpmss_info *info = targinfo; - const struct ipt_entry *e = entry; + const struct xt_tcpmss_info *info = par->targinfo; + const struct ipt_entry *e = par->entryinfo; if (info->mss == XT_TCPMSS_CLAMP_PMTU && - (hook_mask & ~((1 << NF_INET_FORWARD) | + (par->hook_mask & ~((1 << NF_INET_FORWARD) | (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_POST_ROUTING))) != 0) { printk("xt_TCPMSS: path-MTU clamping only supported in " @@ -264,16 +257,13 @@ tcpmss_tg4_check(const char *tablename, const void *entry, } #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) -static bool -tcpmss_tg6_check(const char *tablename, const void *entry, - const struct xt_target *target, void *targinfo, - unsigned int hook_mask) +static bool tcpmss_tg6_check(const struct xt_tgchk_param *par) { - const struct xt_tcpmss_info *info = targinfo; - const struct ip6t_entry *e = entry; + const struct xt_tcpmss_info *info = par->targinfo; + const struct ip6t_entry *e = par->entryinfo; if (info->mss == XT_TCPMSS_CLAMP_PMTU && - (hook_mask & ~((1 << NF_INET_FORWARD) | + (par->hook_mask & ~((1 << NF_INET_FORWARD) | (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_POST_ROUTING))) != 0) { printk("xt_TCPMSS: path-MTU clamping only supported in " @@ -289,7 +279,7 @@ tcpmss_tg6_check(const char *tablename, const void *entry, static struct xt_target tcpmss_tg_reg[] __read_mostly = { { - .family = AF_INET, + .family = NFPROTO_IPV4, .name = "TCPMSS", .checkentry = tcpmss_tg4_check, .target = tcpmss_tg4, @@ -299,7 +289,7 @@ static struct xt_target tcpmss_tg_reg[] __read_mostly = { }, #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) { - .family = AF_INET6, + .family = NFPROTO_IPV6, .name = "TCPMSS", .checkentry = tcpmss_tg6_check, .target = tcpmss_tg6, diff --git a/net/netfilter/xt_TCPOPTSTRIP.c b/net/netfilter/xt_TCPOPTSTRIP.c index 9685b6fcbc8..9dd8c8ef63e 100644 --- a/net/netfilter/xt_TCPOPTSTRIP.c +++ b/net/netfilter/xt_TCPOPTSTRIP.c @@ -75,19 +75,15 @@ tcpoptstrip_mangle_packet(struct sk_buff *skb, } static unsigned int -tcpoptstrip_tg4(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tcpoptstrip_tg4(struct sk_buff *skb, const struct xt_target_param *par) { - return tcpoptstrip_mangle_packet(skb, targinfo, ip_hdrlen(skb), + return tcpoptstrip_mangle_packet(skb, par->targinfo, ip_hdrlen(skb), sizeof(struct iphdr) + sizeof(struct tcphdr)); } #if defined(CONFIG_IP6_NF_MANGLE) || defined(CONFIG_IP6_NF_MANGLE_MODULE) static unsigned int -tcpoptstrip_tg6(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +tcpoptstrip_tg6(struct sk_buff *skb, const struct xt_target_param *par) { struct ipv6hdr *ipv6h = ipv6_hdr(skb); int tcphoff; @@ -98,7 +94,7 @@ tcpoptstrip_tg6(struct sk_buff *skb, const struct net_device *in, if (tcphoff < 0) return NF_DROP; - return tcpoptstrip_mangle_packet(skb, targinfo, tcphoff, + return tcpoptstrip_mangle_packet(skb, par->targinfo, tcphoff, sizeof(*ipv6h) + sizeof(struct tcphdr)); } #endif @@ -106,7 +102,7 @@ tcpoptstrip_tg6(struct sk_buff *skb, const struct net_device *in, static struct xt_target tcpoptstrip_tg_reg[] __read_mostly = { { .name = "TCPOPTSTRIP", - .family = AF_INET, + .family = NFPROTO_IPV4, .table = "mangle", .proto = IPPROTO_TCP, .target = tcpoptstrip_tg4, @@ -116,7 +112,7 @@ static struct xt_target tcpoptstrip_tg_reg[] __read_mostly = { #if defined(CONFIG_IP6_NF_MANGLE) || defined(CONFIG_IP6_NF_MANGLE_MODULE) { .name = "TCPOPTSTRIP", - .family = AF_INET6, + .family = NFPROTO_IPV6, .table = "mangle", .proto = IPPROTO_TCP, .target = tcpoptstrip_tg6, diff --git a/net/netfilter/xt_TPROXY.c b/net/netfilter/xt_TPROXY.c new file mode 100644 index 00000000000..1340c2fa362 --- /dev/null +++ b/net/netfilter/xt_TPROXY.c @@ -0,0 +1,102 @@ +/* + * Transparent proxy support for Linux/iptables + * + * Copyright (c) 2006-2007 BalaBit IT Ltd. + * Author: Balazs Scheidler, Krisztian Kovacs + * + * 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 <net/udp.h> +#include <net/inet_sock.h> + +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter/xt_TPROXY.h> + +#include <net/netfilter/ipv4/nf_defrag_ipv4.h> +#include <net/netfilter/nf_tproxy_core.h> + +static unsigned int +tproxy_tg(struct sk_buff *skb, const struct xt_target_param *par) +{ + const struct iphdr *iph = ip_hdr(skb); + const struct xt_tproxy_target_info *tgi = par->targinfo; + struct udphdr _hdr, *hp; + struct sock *sk; + + hp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(_hdr), &_hdr); + if (hp == NULL) + return NF_DROP; + + sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), iph->protocol, + iph->saddr, tgi->laddr ? tgi->laddr : iph->daddr, + hp->source, tgi->lport ? tgi->lport : hp->dest, + par->in, true); + + /* NOTE: assign_sock consumes our sk reference */ + if (sk && nf_tproxy_assign_sock(skb, sk)) { + /* This should be in a separate target, but we don't do multiple + targets on the same rule yet */ + skb->mark = (skb->mark & ~tgi->mark_mask) ^ tgi->mark_value; + + pr_debug("redirecting: proto %u %08x:%u -> %08x:%u, mark: %x\n", + iph->protocol, ntohl(iph->daddr), ntohs(hp->dest), + ntohl(tgi->laddr), ntohs(tgi->lport), skb->mark); + return NF_ACCEPT; + } + + pr_debug("no socket, dropping: proto %u %08x:%u -> %08x:%u, mark: %x\n", + iph->protocol, ntohl(iph->daddr), ntohs(hp->dest), + ntohl(tgi->laddr), ntohs(tgi->lport), skb->mark); + return NF_DROP; +} + +static bool tproxy_tg_check(const struct xt_tgchk_param *par) +{ + const struct ipt_ip *i = par->entryinfo; + + if ((i->proto == IPPROTO_TCP || i->proto == IPPROTO_UDP) + && !(i->invflags & IPT_INV_PROTO)) + return true; + + pr_info("xt_TPROXY: Can be used only in combination with " + "either -p tcp or -p udp\n"); + return false; +} + +static struct xt_target tproxy_tg_reg __read_mostly = { + .name = "TPROXY", + .family = AF_INET, + .table = "mangle", + .target = tproxy_tg, + .targetsize = sizeof(struct xt_tproxy_target_info), + .checkentry = tproxy_tg_check, + .hooks = 1 << NF_INET_PRE_ROUTING, + .me = THIS_MODULE, +}; + +static int __init tproxy_tg_init(void) +{ + nf_defrag_ipv4_enable(); + return xt_register_target(&tproxy_tg_reg); +} + +static void __exit tproxy_tg_exit(void) +{ + xt_unregister_target(&tproxy_tg_reg); +} + +module_init(tproxy_tg_init); +module_exit(tproxy_tg_exit); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Krisztian Kovacs"); +MODULE_DESCRIPTION("Netfilter transparent proxy (TPROXY) target module."); +MODULE_ALIAS("ipt_TPROXY"); diff --git a/net/netfilter/xt_TRACE.c b/net/netfilter/xt_TRACE.c index 30dab79a343..fbb04b86c46 100644 --- a/net/netfilter/xt_TRACE.c +++ b/net/netfilter/xt_TRACE.c @@ -11,39 +11,29 @@ MODULE_ALIAS("ipt_TRACE"); MODULE_ALIAS("ip6t_TRACE"); static unsigned int -trace_tg(struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, unsigned int hooknum, - const struct xt_target *target, const void *targinfo) +trace_tg(struct sk_buff *skb, const struct xt_target_param *par) { skb->nf_trace = 1; return XT_CONTINUE; } -static struct xt_target trace_tg_reg[] __read_mostly = { - { - .name = "TRACE", - .family = AF_INET, - .target = trace_tg, - .table = "raw", - .me = THIS_MODULE, - }, - { - .name = "TRACE", - .family = AF_INET6, - .target = trace_tg, - .table = "raw", - .me = THIS_MODULE, - }, +static struct xt_target trace_tg_reg __read_mostly = { + .name = "TRACE", + .revision = 0, + .family = NFPROTO_UNSPEC, + .table = "raw", + .target = trace_tg, + .me = THIS_MODULE, }; static int __init trace_tg_init(void) { - return xt_register_targets(trace_tg_reg, ARRAY_SIZE(trace_tg_reg)); + return xt_register_target(&trace_tg_reg); } static void __exit trace_tg_exit(void) { - xt_unregister_targets(trace_tg_reg, ARRAY_SIZE(trace_tg_reg)); + xt_unregister_target(&trace_tg_reg); } module_init(trace_tg_init); diff --git a/net/netfilter/xt_comment.c b/net/netfilter/xt_comment.c index 89f47364e84..e82179832ac 100644 --- a/net/netfilter/xt_comment.c +++ b/net/netfilter/xt_comment.c @@ -16,40 +16,29 @@ MODULE_ALIAS("ipt_comment"); MODULE_ALIAS("ip6t_comment"); static bool -comment_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protooff, - bool *hotdrop) +comment_mt(const struct sk_buff *skb, const struct xt_match_param *par) { /* We always match */ return true; } -static struct xt_match comment_mt_reg[] __read_mostly = { - { - .name = "comment", - .family = AF_INET, - .match = comment_mt, - .matchsize = sizeof(struct xt_comment_info), - .me = THIS_MODULE - }, - { - .name = "comment", - .family = AF_INET6, - .match = comment_mt, - .matchsize = sizeof(struct xt_comment_info), - .me = THIS_MODULE - }, +static struct xt_match comment_mt_reg __read_mostly = { + .name = "comment", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = comment_mt, + .matchsize = sizeof(struct xt_comment_info), + .me = THIS_MODULE, }; static int __init comment_mt_init(void) { - return xt_register_matches(comment_mt_reg, ARRAY_SIZE(comment_mt_reg)); + return xt_register_match(&comment_mt_reg); } static void __exit comment_mt_exit(void) { - xt_unregister_matches(comment_mt_reg, ARRAY_SIZE(comment_mt_reg)); + xt_unregister_match(&comment_mt_reg); } module_init(comment_mt_init); diff --git a/net/netfilter/xt_connbytes.c b/net/netfilter/xt_connbytes.c index 3e39c4fe193..955e6598a7f 100644 --- a/net/netfilter/xt_connbytes.c +++ b/net/netfilter/xt_connbytes.c @@ -17,12 +17,9 @@ MODULE_ALIAS("ipt_connbytes"); MODULE_ALIAS("ip6t_connbytes"); static bool -connbytes_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +connbytes_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_connbytes_info *sinfo = matchinfo; + const struct xt_connbytes_info *sinfo = par->matchinfo; const struct nf_conn *ct; enum ip_conntrack_info ctinfo; u_int64_t what = 0; /* initialize to make gcc happy */ @@ -95,12 +92,9 @@ connbytes_mt(const struct sk_buff *skb, const struct net_device *in, return what >= sinfo->count.from; } -static bool -connbytes_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool connbytes_mt_check(const struct xt_mtchk_param *par) { - const struct xt_connbytes_info *sinfo = matchinfo; + const struct xt_connbytes_info *sinfo = par->matchinfo; if (sinfo->what != XT_CONNBYTES_PKTS && sinfo->what != XT_CONNBYTES_BYTES && @@ -112,51 +106,39 @@ connbytes_mt_check(const char *tablename, const void *ip, sinfo->direction != XT_CONNBYTES_DIR_BOTH) return false; - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", match->family); + "proto=%u\n", par->family); return false; } return true; } -static void -connbytes_mt_destroy(const struct xt_match *match, void *matchinfo) +static void connbytes_mt_destroy(const struct xt_mtdtor_param *par) { - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->family); } -static struct xt_match connbytes_mt_reg[] __read_mostly = { - { - .name = "connbytes", - .family = AF_INET, - .checkentry = connbytes_mt_check, - .match = connbytes_mt, - .destroy = connbytes_mt_destroy, - .matchsize = sizeof(struct xt_connbytes_info), - .me = THIS_MODULE - }, - { - .name = "connbytes", - .family = AF_INET6, - .checkentry = connbytes_mt_check, - .match = connbytes_mt, - .destroy = connbytes_mt_destroy, - .matchsize = sizeof(struct xt_connbytes_info), - .me = THIS_MODULE - }, +static struct xt_match connbytes_mt_reg __read_mostly = { + .name = "connbytes", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = connbytes_mt_check, + .match = connbytes_mt, + .destroy = connbytes_mt_destroy, + .matchsize = sizeof(struct xt_connbytes_info), + .me = THIS_MODULE, }; static int __init connbytes_mt_init(void) { - return xt_register_matches(connbytes_mt_reg, - ARRAY_SIZE(connbytes_mt_reg)); + return xt_register_match(&connbytes_mt_reg); } static void __exit connbytes_mt_exit(void) { - xt_unregister_matches(connbytes_mt_reg, ARRAY_SIZE(connbytes_mt_reg)); + xt_unregister_match(&connbytes_mt_reg); } module_init(connbytes_mt_init); diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 70907f6baac..7f404cc64c8 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -82,9 +82,9 @@ static inline bool already_closed(const struct nf_conn *conn) static inline unsigned int same_source_net(const union nf_inet_addr *addr, const union nf_inet_addr *mask, - const union nf_inet_addr *u3, unsigned int family) + const union nf_inet_addr *u3, u_int8_t family) { - if (family == AF_INET) { + if (family == NFPROTO_IPV4) { return (addr->ip & mask->ip) == (u3->ip & mask->ip); } else { union nf_inet_addr lh, rh; @@ -114,7 +114,7 @@ static int count_them(struct xt_connlimit_data *data, int matches = 0; - if (match->family == AF_INET6) + if (match->family == NFPROTO_IPV6) hash = &data->iphash[connlimit_iphash6(addr, mask)]; else hash = &data->iphash[connlimit_iphash(addr->ip & mask->ip)]; @@ -123,7 +123,7 @@ static int count_them(struct xt_connlimit_data *data, /* check the saved connections */ list_for_each_entry_safe(conn, tmp, hash, list) { - found = __nf_conntrack_find(&conn->tuple); + found = __nf_conntrack_find(&init_net, &conn->tuple); found_ct = NULL; if (found != NULL) @@ -178,12 +178,9 @@ static int count_them(struct xt_connlimit_data *data, } static bool -connlimit_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +connlimit_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_connlimit_info *info = matchinfo; + const struct xt_connlimit_info *info = par->matchinfo; union nf_inet_addr addr; struct nf_conntrack_tuple tuple; const struct nf_conntrack_tuple *tuple_ptr = &tuple; @@ -195,10 +192,10 @@ connlimit_mt(const struct sk_buff *skb, const struct net_device *in, if (ct != NULL) tuple_ptr = &ct->tuplehash[0].tuple; else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), - match->family, &tuple)) + par->family, &tuple)) goto hotdrop; - if (match->family == AF_INET6) { + if (par->family == NFPROTO_IPV6) { const struct ipv6hdr *iph = ipv6_hdr(skb); memcpy(&addr.ip6, &iph->saddr, sizeof(iph->saddr)); } else { @@ -208,40 +205,37 @@ connlimit_mt(const struct sk_buff *skb, const struct net_device *in, spin_lock_bh(&info->data->lock); connections = count_them(info->data, tuple_ptr, &addr, - &info->mask, match); + &info->mask, par->match); spin_unlock_bh(&info->data->lock); if (connections < 0) { /* kmalloc failed, drop it entirely */ - *hotdrop = true; + *par->hotdrop = true; return false; } return (connections > info->limit) ^ info->inverse; hotdrop: - *hotdrop = true; + *par->hotdrop = true; return false; } -static bool -connlimit_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool connlimit_mt_check(const struct xt_mtchk_param *par) { - struct xt_connlimit_info *info = matchinfo; + struct xt_connlimit_info *info = par->matchinfo; unsigned int i; - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "cannot load conntrack support for " - "address family %u\n", match->family); + "address family %u\n", par->family); return false; } /* init private data */ info->data = kmalloc(sizeof(struct xt_connlimit_data), GFP_KERNEL); if (info->data == NULL) { - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->family); return false; } @@ -252,16 +246,15 @@ connlimit_mt_check(const char *tablename, const void *ip, return true; } -static void -connlimit_mt_destroy(const struct xt_match *match, void *matchinfo) +static void connlimit_mt_destroy(const struct xt_mtdtor_param *par) { - const struct xt_connlimit_info *info = matchinfo; + const struct xt_connlimit_info *info = par->matchinfo; struct xt_connlimit_conn *conn; struct xt_connlimit_conn *tmp; struct list_head *hash = info->data->iphash; unsigned int i; - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->family); for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i) { list_for_each_entry_safe(conn, tmp, &hash[i], list) { @@ -273,41 +266,30 @@ connlimit_mt_destroy(const struct xt_match *match, void *matchinfo) kfree(info->data); } -static struct xt_match connlimit_mt_reg[] __read_mostly = { - { - .name = "connlimit", - .family = AF_INET, - .checkentry = connlimit_mt_check, - .match = connlimit_mt, - .matchsize = sizeof(struct xt_connlimit_info), - .destroy = connlimit_mt_destroy, - .me = THIS_MODULE, - }, - { - .name = "connlimit", - .family = AF_INET6, - .checkentry = connlimit_mt_check, - .match = connlimit_mt, - .matchsize = sizeof(struct xt_connlimit_info), - .destroy = connlimit_mt_destroy, - .me = THIS_MODULE, - }, +static struct xt_match connlimit_mt_reg __read_mostly = { + .name = "connlimit", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = connlimit_mt_check, + .match = connlimit_mt, + .matchsize = sizeof(struct xt_connlimit_info), + .destroy = connlimit_mt_destroy, + .me = THIS_MODULE, }; static int __init connlimit_mt_init(void) { - return xt_register_matches(connlimit_mt_reg, - ARRAY_SIZE(connlimit_mt_reg)); + return xt_register_match(&connlimit_mt_reg); } static void __exit connlimit_mt_exit(void) { - xt_unregister_matches(connlimit_mt_reg, ARRAY_SIZE(connlimit_mt_reg)); + xt_unregister_match(&connlimit_mt_reg); } module_init(connlimit_mt_init); module_exit(connlimit_mt_exit); -MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>"); +MODULE_AUTHOR("Jan Engelhardt <jengelh@medozas.de>"); MODULE_DESCRIPTION("Xtables: Number of connections matching"); MODULE_LICENSE("GPL"); MODULE_ALIAS("ipt_connlimit"); diff --git a/net/netfilter/xt_connmark.c b/net/netfilter/xt_connmark.c index aaa1b96691f..86cacab7a4a 100644 --- a/net/netfilter/xt_connmark.c +++ b/net/netfilter/xt_connmark.c @@ -34,12 +34,9 @@ MODULE_ALIAS("ipt_connmark"); MODULE_ALIAS("ip6t_connmark"); static bool -connmark_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +connmark_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_connmark_mtinfo1 *info = matchinfo; + const struct xt_connmark_mtinfo1 *info = par->matchinfo; enum ip_conntrack_info ctinfo; const struct nf_conn *ct; @@ -51,12 +48,9 @@ connmark_mt(const struct sk_buff *skb, const struct net_device *in, } static bool -connmark_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +connmark_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_connmark_info *info = matchinfo; + const struct xt_connmark_info *info = par->matchinfo; const struct nf_conn *ct; enum ip_conntrack_info ctinfo; @@ -67,42 +61,35 @@ connmark_mt_v0(const struct sk_buff *skb, const struct net_device *in, return ((ct->mark & info->mask) == info->mark) ^ info->invert; } -static bool -connmark_mt_check_v0(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool connmark_mt_check_v0(const struct xt_mtchk_param *par) { - const struct xt_connmark_info *cm = matchinfo; + const struct xt_connmark_info *cm = par->matchinfo; if (cm->mark > 0xffffffff || cm->mask > 0xffffffff) { printk(KERN_WARNING "connmark: only support 32bit mark\n"); return false; } - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", match->family); + "proto=%u\n", par->family); return false; } return true; } -static bool -connmark_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool connmark_mt_check(const struct xt_mtchk_param *par) { - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "cannot load conntrack support for " - "proto=%u\n", match->family); + "proto=%u\n", par->family); return false; } return true; } -static void -connmark_mt_destroy(const struct xt_match *match, void *matchinfo) +static void connmark_mt_destroy(const struct xt_mtdtor_param *par) { - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->family); } #ifdef CONFIG_COMPAT @@ -140,7 +127,7 @@ static struct xt_match connmark_mt_reg[] __read_mostly = { { .name = "connmark", .revision = 0, - .family = AF_INET, + .family = NFPROTO_UNSPEC, .checkentry = connmark_mt_check_v0, .match = connmark_mt_v0, .destroy = connmark_mt_destroy, @@ -153,34 +140,9 @@ static struct xt_match connmark_mt_reg[] __read_mostly = { .me = THIS_MODULE }, { - .name = "connmark", - .revision = 0, - .family = AF_INET6, - .checkentry = connmark_mt_check_v0, - .match = connmark_mt_v0, - .destroy = connmark_mt_destroy, - .matchsize = sizeof(struct xt_connmark_info), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_connmark_info), - .compat_from_user = connmark_mt_compat_from_user_v0, - .compat_to_user = connmark_mt_compat_to_user_v0, -#endif - .me = THIS_MODULE - }, - { - .name = "connmark", - .revision = 1, - .family = AF_INET, - .checkentry = connmark_mt_check, - .match = connmark_mt, - .matchsize = sizeof(struct xt_connmark_mtinfo1), - .destroy = connmark_mt_destroy, - .me = THIS_MODULE, - }, - { .name = "connmark", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .checkentry = connmark_mt_check, .match = connmark_mt, .matchsize = sizeof(struct xt_connmark_mtinfo1), diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c index d61412f58ef..0b7139f3dd7 100644 --- a/net/netfilter/xt_conntrack.c +++ b/net/netfilter/xt_conntrack.c @@ -25,12 +25,9 @@ MODULE_ALIAS("ipt_conntrack"); MODULE_ALIAS("ip6t_conntrack"); static bool -conntrack_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +conntrack_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_conntrack_info *sinfo = matchinfo; + const struct xt_conntrack_info *sinfo = par->matchinfo; const struct nf_conn *ct; enum ip_conntrack_info ctinfo; unsigned int statebit; @@ -121,9 +118,9 @@ conntrack_addrcmp(const union nf_inet_addr *kaddr, const union nf_inet_addr *uaddr, const union nf_inet_addr *umask, unsigned int l3proto) { - if (l3proto == AF_INET) + if (l3proto == NFPROTO_IPV4) return ((kaddr->ip ^ uaddr->ip) & umask->ip) == 0; - else if (l3proto == AF_INET6) + else if (l3proto == NFPROTO_IPV6) return ipv6_masked_addr_cmp(&kaddr->in6, &umask->in6, &uaddr->in6) == 0; else @@ -133,7 +130,7 @@ conntrack_addrcmp(const union nf_inet_addr *kaddr, static inline bool conntrack_mt_origsrc(const struct nf_conn *ct, const struct xt_conntrack_mtinfo1 *info, - unsigned int family) + u_int8_t family) { return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3, &info->origsrc_addr, &info->origsrc_mask, family); @@ -142,7 +139,7 @@ conntrack_mt_origsrc(const struct nf_conn *ct, static inline bool conntrack_mt_origdst(const struct nf_conn *ct, const struct xt_conntrack_mtinfo1 *info, - unsigned int family) + u_int8_t family) { return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3, &info->origdst_addr, &info->origdst_mask, family); @@ -151,7 +148,7 @@ conntrack_mt_origdst(const struct nf_conn *ct, static inline bool conntrack_mt_replsrc(const struct nf_conn *ct, const struct xt_conntrack_mtinfo1 *info, - unsigned int family) + u_int8_t family) { return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3, &info->replsrc_addr, &info->replsrc_mask, family); @@ -160,7 +157,7 @@ conntrack_mt_replsrc(const struct nf_conn *ct, static inline bool conntrack_mt_repldst(const struct nf_conn *ct, const struct xt_conntrack_mtinfo1 *info, - unsigned int family) + u_int8_t family) { return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3, &info->repldst_addr, &info->repldst_mask, family); @@ -205,12 +202,9 @@ ct_proto_port_check(const struct xt_conntrack_mtinfo1 *info, } static bool -conntrack_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +conntrack_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_conntrack_mtinfo1 *info = matchinfo; + const struct xt_conntrack_mtinfo1 *info = par->matchinfo; enum ip_conntrack_info ctinfo; const struct nf_conn *ct; unsigned int statebit; @@ -244,22 +238,22 @@ conntrack_mt(const struct sk_buff *skb, const struct net_device *in, return false; if (info->match_flags & XT_CONNTRACK_ORIGSRC) - if (conntrack_mt_origsrc(ct, info, match->family) ^ + if (conntrack_mt_origsrc(ct, info, par->family) ^ !(info->invert_flags & XT_CONNTRACK_ORIGSRC)) return false; if (info->match_flags & XT_CONNTRACK_ORIGDST) - if (conntrack_mt_origdst(ct, info, match->family) ^ + if (conntrack_mt_origdst(ct, info, par->family) ^ !(info->invert_flags & XT_CONNTRACK_ORIGDST)) return false; if (info->match_flags & XT_CONNTRACK_REPLSRC) - if (conntrack_mt_replsrc(ct, info, match->family) ^ + if (conntrack_mt_replsrc(ct, info, par->family) ^ !(info->invert_flags & XT_CONNTRACK_REPLSRC)) return false; if (info->match_flags & XT_CONNTRACK_REPLDST) - if (conntrack_mt_repldst(ct, info, match->family) ^ + if (conntrack_mt_repldst(ct, info, par->family) ^ !(info->invert_flags & XT_CONNTRACK_REPLDST)) return false; @@ -284,23 +278,19 @@ conntrack_mt(const struct sk_buff *skb, const struct net_device *in, return true; } -static bool -conntrack_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool conntrack_mt_check(const struct xt_mtchk_param *par) { - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", match->family); + "proto=%u\n", par->family); return false; } return true; } -static void -conntrack_mt_destroy(const struct xt_match *match, void *matchinfo) +static void conntrack_mt_destroy(const struct xt_mtdtor_param *par) { - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->family); } #ifdef CONFIG_COMPAT @@ -356,7 +346,7 @@ static struct xt_match conntrack_mt_reg[] __read_mostly = { { .name = "conntrack", .revision = 0, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = conntrack_mt_v0, .checkentry = conntrack_mt_check, .destroy = conntrack_mt_destroy, @@ -371,17 +361,7 @@ static struct xt_match conntrack_mt_reg[] __read_mostly = { { .name = "conntrack", .revision = 1, - .family = AF_INET, - .matchsize = sizeof(struct xt_conntrack_mtinfo1), - .match = conntrack_mt, - .checkentry = conntrack_mt_check, - .destroy = conntrack_mt_destroy, - .me = THIS_MODULE, - }, - { - .name = "conntrack", - .revision = 1, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .matchsize = sizeof(struct xt_conntrack_mtinfo1), .match = conntrack_mt, .checkentry = conntrack_mt_check, diff --git a/net/netfilter/xt_dccp.c b/net/netfilter/xt_dccp.c index 8b6522186d9..e5d3e867328 100644 --- a/net/netfilter/xt_dccp.c +++ b/net/netfilter/xt_dccp.c @@ -93,20 +93,18 @@ match_option(u_int8_t option, const struct sk_buff *skb, unsigned int protoff, } static bool -dccp_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +dccp_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_dccp_info *info = matchinfo; + const struct xt_dccp_info *info = par->matchinfo; const struct dccp_hdr *dh; struct dccp_hdr _dh; - if (offset) + if (par->fragoff != 0) return false; - dh = skb_header_pointer(skb, protoff, sizeof(_dh), &_dh); + dh = skb_header_pointer(skb, par->thoff, sizeof(_dh), &_dh); if (dh == NULL) { - *hotdrop = true; + *par->hotdrop = true; return false; } @@ -118,17 +116,14 @@ dccp_mt(const struct sk_buff *skb, const struct net_device *in, XT_DCCP_DEST_PORTS, info->flags, info->invflags) && DCCHECK(match_types(dh, info->typemask), XT_DCCP_TYPE, info->flags, info->invflags) - && DCCHECK(match_option(info->option, skb, protoff, dh, - hotdrop), + && DCCHECK(match_option(info->option, skb, par->thoff, dh, + par->hotdrop), XT_DCCP_OPTION, info->flags, info->invflags); } -static bool -dccp_mt_check(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool dccp_mt_check(const struct xt_mtchk_param *par) { - const struct xt_dccp_info *info = matchinfo; + const struct xt_dccp_info *info = par->matchinfo; return !(info->flags & ~XT_DCCP_VALID_FLAGS) && !(info->invflags & ~XT_DCCP_VALID_FLAGS) @@ -138,7 +133,7 @@ dccp_mt_check(const char *tablename, const void *inf, static struct xt_match dccp_mt_reg[] __read_mostly = { { .name = "dccp", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = dccp_mt_check, .match = dccp_mt, .matchsize = sizeof(struct xt_dccp_info), @@ -147,7 +142,7 @@ static struct xt_match dccp_mt_reg[] __read_mostly = { }, { .name = "dccp", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = dccp_mt_check, .match = dccp_mt, .matchsize = sizeof(struct xt_dccp_info), diff --git a/net/netfilter/xt_dscp.c b/net/netfilter/xt_dscp.c index 26f4aab9c42..c3f8085460d 100644 --- a/net/netfilter/xt_dscp.c +++ b/net/netfilter/xt_dscp.c @@ -26,61 +26,48 @@ MODULE_ALIAS("ipt_tos"); MODULE_ALIAS("ip6t_tos"); static bool -dscp_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +dscp_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_dscp_info *info = matchinfo; + const struct xt_dscp_info *info = par->matchinfo; u_int8_t dscp = ipv4_get_dsfield(ip_hdr(skb)) >> XT_DSCP_SHIFT; return (dscp == info->dscp) ^ !!info->invert; } static bool -dscp_mt6(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +dscp_mt6(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_dscp_info *info = matchinfo; + const struct xt_dscp_info *info = par->matchinfo; u_int8_t dscp = ipv6_get_dsfield(ipv6_hdr(skb)) >> XT_DSCP_SHIFT; return (dscp == info->dscp) ^ !!info->invert; } -static bool -dscp_mt_check(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool dscp_mt_check(const struct xt_mtchk_param *par) { - const u_int8_t dscp = ((struct xt_dscp_info *)matchinfo)->dscp; + const struct xt_dscp_info *info = par->matchinfo; - if (dscp > XT_DSCP_MAX) { - printk(KERN_ERR "xt_dscp: dscp %x out of range\n", dscp); + if (info->dscp > XT_DSCP_MAX) { + printk(KERN_ERR "xt_dscp: dscp %x out of range\n", info->dscp); return false; } return true; } -static bool tos_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, - const struct xt_match *match, const void *matchinfo, - int offset, unsigned int protoff, bool *hotdrop) +static bool +tos_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct ipt_tos_info *info = matchinfo; + const struct ipt_tos_info *info = par->matchinfo; return (ip_hdr(skb)->tos == info->tos) ^ info->invert; } -static bool tos_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +static bool tos_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_tos_match_info *info = matchinfo; + const struct xt_tos_match_info *info = par->matchinfo; - if (match->family == AF_INET) + if (par->match->family == NFPROTO_IPV4) return ((ip_hdr(skb)->tos & info->tos_mask) == info->tos_value) ^ !!info->invert; else @@ -91,7 +78,7 @@ static bool tos_mt(const struct sk_buff *skb, const struct net_device *in, static struct xt_match dscp_mt_reg[] __read_mostly = { { .name = "dscp", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = dscp_mt_check, .match = dscp_mt, .matchsize = sizeof(struct xt_dscp_info), @@ -99,7 +86,7 @@ static struct xt_match dscp_mt_reg[] __read_mostly = { }, { .name = "dscp", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = dscp_mt_check, .match = dscp_mt6, .matchsize = sizeof(struct xt_dscp_info), @@ -108,7 +95,7 @@ static struct xt_match dscp_mt_reg[] __read_mostly = { { .name = "tos", .revision = 0, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = tos_mt_v0, .matchsize = sizeof(struct ipt_tos_info), .me = THIS_MODULE, @@ -116,7 +103,7 @@ static struct xt_match dscp_mt_reg[] __read_mostly = { { .name = "tos", .revision = 1, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = tos_mt, .matchsize = sizeof(struct xt_tos_match_info), .me = THIS_MODULE, @@ -124,7 +111,7 @@ static struct xt_match dscp_mt_reg[] __read_mostly = { { .name = "tos", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = tos_mt, .matchsize = sizeof(struct xt_tos_match_info), .me = THIS_MODULE, diff --git a/net/netfilter/xt_esp.c b/net/netfilter/xt_esp.c index a133eb9b23e..609439967c2 100644 --- a/net/netfilter/xt_esp.c +++ b/net/netfilter/xt_esp.c @@ -42,26 +42,23 @@ spi_match(u_int32_t min, u_int32_t max, u_int32_t spi, bool invert) return r; } -static bool -esp_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +static bool esp_mt(const struct sk_buff *skb, const struct xt_match_param *par) { const struct ip_esp_hdr *eh; struct ip_esp_hdr _esp; - const struct xt_esp *espinfo = matchinfo; + const struct xt_esp *espinfo = par->matchinfo; /* Must not be a fragment. */ - if (offset) + if (par->fragoff != 0) return false; - eh = skb_header_pointer(skb, protoff, sizeof(_esp), &_esp); + eh = skb_header_pointer(skb, par->thoff, sizeof(_esp), &_esp); if (eh == NULL) { /* We've been asked to examine this packet, and we * can't. Hence, no choice but to drop. */ duprintf("Dropping evil ESP tinygram.\n"); - *hotdrop = true; + *par->hotdrop = true; return false; } @@ -69,13 +66,9 @@ esp_mt(const struct sk_buff *skb, const struct net_device *in, !!(espinfo->invflags & XT_ESP_INV_SPI)); } -/* Called when user tries to insert an entry of this type. */ -static bool -esp_mt_check(const char *tablename, const void *ip_void, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool esp_mt_check(const struct xt_mtchk_param *par) { - const struct xt_esp *espinfo = matchinfo; + const struct xt_esp *espinfo = par->matchinfo; if (espinfo->invflags & ~XT_ESP_INV_MASK) { duprintf("xt_esp: unknown flags %X\n", espinfo->invflags); @@ -88,7 +81,7 @@ esp_mt_check(const char *tablename, const void *ip_void, static struct xt_match esp_mt_reg[] __read_mostly = { { .name = "esp", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = esp_mt_check, .match = esp_mt, .matchsize = sizeof(struct xt_esp), @@ -97,7 +90,7 @@ static struct xt_match esp_mt_reg[] __read_mostly = { }, { .name = "esp", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = esp_mt_check, .match = esp_mt, .matchsize = sizeof(struct xt_esp), diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c index d9418a26781..6fc4292d46e 100644 --- a/net/netfilter/xt_hashlimit.c +++ b/net/netfilter/xt_hashlimit.c @@ -80,7 +80,7 @@ struct dsthash_ent { struct xt_hashlimit_htable { struct hlist_node node; /* global list of all htables */ atomic_t use; - int family; + u_int8_t family; struct hashlimit_cfg1 cfg; /* config */ @@ -185,7 +185,7 @@ dsthash_free(struct xt_hashlimit_htable *ht, struct dsthash_ent *ent) } static void htable_gc(unsigned long htlong); -static int htable_create_v0(struct xt_hashlimit_info *minfo, int family) +static int htable_create_v0(struct xt_hashlimit_info *minfo, u_int8_t family) { struct xt_hashlimit_htable *hinfo; unsigned int size; @@ -218,7 +218,7 @@ static int htable_create_v0(struct xt_hashlimit_info *minfo, int family) hinfo->cfg.gc_interval = minfo->cfg.gc_interval; hinfo->cfg.expire = minfo->cfg.expire; - if (family == AF_INET) + if (family == NFPROTO_IPV4) hinfo->cfg.srcmask = hinfo->cfg.dstmask = 32; else hinfo->cfg.srcmask = hinfo->cfg.dstmask = 128; @@ -237,11 +237,10 @@ static int htable_create_v0(struct xt_hashlimit_info *minfo, int family) hinfo->family = family; hinfo->rnd_initialized = 0; spin_lock_init(&hinfo->lock); - hinfo->pde = - proc_create_data(minfo->name, 0, - family == AF_INET ? hashlimit_procdir4 : - hashlimit_procdir6, - &dl_file_ops, hinfo); + hinfo->pde = proc_create_data(minfo->name, 0, + (family == NFPROTO_IPV4) ? + hashlimit_procdir4 : hashlimit_procdir6, + &dl_file_ops, hinfo); if (!hinfo->pde) { vfree(hinfo); return -1; @@ -258,8 +257,7 @@ static int htable_create_v0(struct xt_hashlimit_info *minfo, int family) return 0; } -static int htable_create(struct xt_hashlimit_mtinfo1 *minfo, - unsigned int family) +static int htable_create(struct xt_hashlimit_mtinfo1 *minfo, u_int8_t family) { struct xt_hashlimit_htable *hinfo; unsigned int size; @@ -301,11 +299,10 @@ static int htable_create(struct xt_hashlimit_mtinfo1 *minfo, hinfo->rnd_initialized = 0; spin_lock_init(&hinfo->lock); - hinfo->pde = - proc_create_data(minfo->name, 0, - family == AF_INET ? hashlimit_procdir4 : - hashlimit_procdir6, - &dl_file_ops, hinfo); + hinfo->pde = proc_create_data(minfo->name, 0, + (family == NFPROTO_IPV4) ? + hashlimit_procdir4 : hashlimit_procdir6, + &dl_file_ops, hinfo); if (hinfo->pde == NULL) { vfree(hinfo); return -1; @@ -371,14 +368,14 @@ static void htable_destroy(struct xt_hashlimit_htable *hinfo) /* remove proc entry */ remove_proc_entry(hinfo->pde->name, - hinfo->family == AF_INET ? hashlimit_procdir4 : + hinfo->family == NFPROTO_IPV4 ? hashlimit_procdir4 : hashlimit_procdir6); htable_selective_cleanup(hinfo, select_all); vfree(hinfo); } static struct xt_hashlimit_htable *htable_find_get(const char *name, - int family) + u_int8_t family) { struct xt_hashlimit_htable *hinfo; struct hlist_node *pos; @@ -502,7 +499,7 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, memset(dst, 0, sizeof(*dst)); switch (hinfo->family) { - case AF_INET: + case NFPROTO_IPV4: if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) dst->ip.dst = maskl(ip_hdr(skb)->daddr, hinfo->cfg.dstmask); @@ -516,7 +513,7 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, nexthdr = ip_hdr(skb)->protocol; break; #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) - case AF_INET6: + case NFPROTO_IPV6: if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) { memcpy(&dst->ip6.dst, &ipv6_hdr(skb)->daddr, sizeof(dst->ip6.dst)); @@ -566,19 +563,16 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo, } static bool -hashlimit_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +hashlimit_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { const struct xt_hashlimit_info *r = - ((const struct xt_hashlimit_info *)matchinfo)->u.master; + ((const struct xt_hashlimit_info *)par->matchinfo)->u.master; struct xt_hashlimit_htable *hinfo = r->hinfo; unsigned long now = jiffies; struct dsthash_ent *dh; struct dsthash_dst dst; - if (hashlimit_init_dst(hinfo, &dst, skb, protoff) < 0) + if (hashlimit_init_dst(hinfo, &dst, skb, par->thoff) < 0) goto hotdrop; spin_lock_bh(&hinfo->lock); @@ -616,23 +610,20 @@ hashlimit_mt_v0(const struct sk_buff *skb, const struct net_device *in, return false; hotdrop: - *hotdrop = true; + *par->hotdrop = true; return false; } static bool -hashlimit_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +hashlimit_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_hashlimit_mtinfo1 *info = matchinfo; + const struct xt_hashlimit_mtinfo1 *info = par->matchinfo; struct xt_hashlimit_htable *hinfo = info->hinfo; unsigned long now = jiffies; struct dsthash_ent *dh; struct dsthash_dst dst; - if (hashlimit_init_dst(hinfo, &dst, skb, protoff) < 0) + if (hashlimit_init_dst(hinfo, &dst, skb, par->thoff) < 0) goto hotdrop; spin_lock_bh(&hinfo->lock); @@ -669,16 +660,13 @@ hashlimit_mt(const struct sk_buff *skb, const struct net_device *in, return info->cfg.mode & XT_HASHLIMIT_INVERT; hotdrop: - *hotdrop = true; + *par->hotdrop = true; return false; } -static bool -hashlimit_mt_check_v0(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool hashlimit_mt_check_v0(const struct xt_mtchk_param *par) { - struct xt_hashlimit_info *r = matchinfo; + struct xt_hashlimit_info *r = par->matchinfo; /* Check for overflow. */ if (r->cfg.burst == 0 || @@ -707,8 +695,8 @@ hashlimit_mt_check_v0(const char *tablename, const void *inf, * the list of htable's in htable_create(), since then we would * create duplicate proc files. -HW */ mutex_lock(&hlimit_mutex); - r->hinfo = htable_find_get(r->name, match->family); - if (!r->hinfo && htable_create_v0(r, match->family) != 0) { + r->hinfo = htable_find_get(r->name, par->match->family); + if (!r->hinfo && htable_create_v0(r, par->match->family) != 0) { mutex_unlock(&hlimit_mutex); return false; } @@ -719,12 +707,9 @@ hashlimit_mt_check_v0(const char *tablename, const void *inf, return true; } -static bool -hashlimit_mt_check(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool hashlimit_mt_check(const struct xt_mtchk_param *par) { - struct xt_hashlimit_mtinfo1 *info = matchinfo; + struct xt_hashlimit_mtinfo1 *info = par->matchinfo; /* Check for overflow. */ if (info->cfg.burst == 0 || @@ -738,7 +723,7 @@ hashlimit_mt_check(const char *tablename, const void *inf, return false; if (info->name[sizeof(info->name)-1] != '\0') return false; - if (match->family == AF_INET) { + if (par->match->family == NFPROTO_IPV4) { if (info->cfg.srcmask > 32 || info->cfg.dstmask > 32) return false; } else { @@ -753,8 +738,8 @@ hashlimit_mt_check(const char *tablename, const void *inf, * the list of htable's in htable_create(), since then we would * create duplicate proc files. -HW */ mutex_lock(&hlimit_mutex); - info->hinfo = htable_find_get(info->name, match->family); - if (!info->hinfo && htable_create(info, match->family) != 0) { + info->hinfo = htable_find_get(info->name, par->match->family); + if (!info->hinfo && htable_create(info, par->match->family) != 0) { mutex_unlock(&hlimit_mutex); return false; } @@ -763,17 +748,16 @@ hashlimit_mt_check(const char *tablename, const void *inf, } static void -hashlimit_mt_destroy_v0(const struct xt_match *match, void *matchinfo) +hashlimit_mt_destroy_v0(const struct xt_mtdtor_param *par) { - const struct xt_hashlimit_info *r = matchinfo; + const struct xt_hashlimit_info *r = par->matchinfo; htable_put(r->hinfo); } -static void -hashlimit_mt_destroy(const struct xt_match *match, void *matchinfo) +static void hashlimit_mt_destroy(const struct xt_mtdtor_param *par) { - const struct xt_hashlimit_mtinfo1 *info = matchinfo; + const struct xt_hashlimit_mtinfo1 *info = par->matchinfo; htable_put(info->hinfo); } @@ -806,7 +790,7 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = { { .name = "hashlimit", .revision = 0, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = hashlimit_mt_v0, .matchsize = sizeof(struct xt_hashlimit_info), #ifdef CONFIG_COMPAT @@ -821,7 +805,7 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = { { .name = "hashlimit", .revision = 1, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = hashlimit_mt, .matchsize = sizeof(struct xt_hashlimit_mtinfo1), .checkentry = hashlimit_mt_check, @@ -831,7 +815,7 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = { #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) { .name = "hashlimit", - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = hashlimit_mt_v0, .matchsize = sizeof(struct xt_hashlimit_info), #ifdef CONFIG_COMPAT @@ -846,7 +830,7 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = { { .name = "hashlimit", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = hashlimit_mt, .matchsize = sizeof(struct xt_hashlimit_mtinfo1), .checkentry = hashlimit_mt_check, @@ -901,14 +885,14 @@ static void dl_seq_stop(struct seq_file *s, void *v) spin_unlock_bh(&htable->lock); } -static int dl_seq_real_show(struct dsthash_ent *ent, int family, +static int dl_seq_real_show(struct dsthash_ent *ent, u_int8_t family, struct seq_file *s) { /* recalculate to show accurate numbers */ rateinfo_recalc(ent, jiffies); switch (family) { - case AF_INET: + case NFPROTO_IPV4: return seq_printf(s, "%ld %u.%u.%u.%u:%u->" "%u.%u.%u.%u:%u %u %u %u\n", (long)(ent->expires - jiffies)/HZ, @@ -919,7 +903,7 @@ static int dl_seq_real_show(struct dsthash_ent *ent, int family, ent->rateinfo.credit, ent->rateinfo.credit_cap, ent->rateinfo.cost); #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) - case AF_INET6: + case NFPROTO_IPV6: return seq_printf(s, "%ld " NIP6_FMT ":%u->" NIP6_FMT ":%u %u %u %u\n", (long)(ent->expires - jiffies)/HZ, diff --git a/net/netfilter/xt_helper.c b/net/netfilter/xt_helper.c index dada2905d66..64fc7f27722 100644 --- a/net/netfilter/xt_helper.c +++ b/net/netfilter/xt_helper.c @@ -24,12 +24,9 @@ MODULE_ALIAS("ip6t_helper"); static bool -helper_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +helper_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_helper_info *info = matchinfo; + const struct xt_helper_info *info = par->matchinfo; const struct nf_conn *ct; const struct nf_conn_help *master_help; const struct nf_conntrack_helper *helper; @@ -57,56 +54,43 @@ helper_mt(const struct sk_buff *skb, const struct net_device *in, return ret; } -static bool -helper_mt_check(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool helper_mt_check(const struct xt_mtchk_param *par) { - struct xt_helper_info *info = matchinfo; + struct xt_helper_info *info = par->matchinfo; - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", match->family); + "proto=%u\n", par->family); return false; } info->name[29] = '\0'; return true; } -static void helper_mt_destroy(const struct xt_match *match, void *matchinfo) +static void helper_mt_destroy(const struct xt_mtdtor_param *par) { - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->family); } -static struct xt_match helper_mt_reg[] __read_mostly = { - { - .name = "helper", - .family = AF_INET, - .checkentry = helper_mt_check, - .match = helper_mt, - .destroy = helper_mt_destroy, - .matchsize = sizeof(struct xt_helper_info), - .me = THIS_MODULE, - }, - { - .name = "helper", - .family = AF_INET6, - .checkentry = helper_mt_check, - .match = helper_mt, - .destroy = helper_mt_destroy, - .matchsize = sizeof(struct xt_helper_info), - .me = THIS_MODULE, - }, +static struct xt_match helper_mt_reg __read_mostly = { + .name = "helper", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = helper_mt_check, + .match = helper_mt, + .destroy = helper_mt_destroy, + .matchsize = sizeof(struct xt_helper_info), + .me = THIS_MODULE, }; static int __init helper_mt_init(void) { - return xt_register_matches(helper_mt_reg, ARRAY_SIZE(helper_mt_reg)); + return xt_register_match(&helper_mt_reg); } static void __exit helper_mt_exit(void) { - xt_unregister_matches(helper_mt_reg, ARRAY_SIZE(helper_mt_reg)); + xt_unregister_match(&helper_mt_reg); } module_init(helper_mt_init); diff --git a/net/netfilter/xt_iprange.c b/net/netfilter/xt_iprange.c index c63e9333c75..6f62c36948d 100644 --- a/net/netfilter/xt_iprange.c +++ b/net/netfilter/xt_iprange.c @@ -17,12 +17,9 @@ #include <linux/netfilter_ipv4/ipt_iprange.h> static bool -iprange_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +iprange_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct ipt_iprange_info *info = matchinfo; + const struct ipt_iprange_info *info = par->matchinfo; const struct iphdr *iph = ip_hdr(skb); if (info->flags & IPRANGE_SRC) { @@ -55,12 +52,9 @@ iprange_mt_v0(const struct sk_buff *skb, const struct net_device *in, } static bool -iprange_mt4(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +iprange_mt4(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_iprange_mtinfo *info = matchinfo; + const struct xt_iprange_mtinfo *info = par->matchinfo; const struct iphdr *iph = ip_hdr(skb); bool m; @@ -111,12 +105,9 @@ iprange_ipv6_sub(const struct in6_addr *a, const struct in6_addr *b) } static bool -iprange_mt6(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +iprange_mt6(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_iprange_mtinfo *info = matchinfo; + const struct xt_iprange_mtinfo *info = par->matchinfo; const struct ipv6hdr *iph = ipv6_hdr(skb); bool m; @@ -141,7 +132,7 @@ static struct xt_match iprange_mt_reg[] __read_mostly = { { .name = "iprange", .revision = 0, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = iprange_mt_v0, .matchsize = sizeof(struct ipt_iprange_info), .me = THIS_MODULE, @@ -149,7 +140,7 @@ static struct xt_match iprange_mt_reg[] __read_mostly = { { .name = "iprange", .revision = 1, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = iprange_mt4, .matchsize = sizeof(struct xt_iprange_mtinfo), .me = THIS_MODULE, @@ -157,7 +148,7 @@ static struct xt_match iprange_mt_reg[] __read_mostly = { { .name = "iprange", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = iprange_mt6, .matchsize = sizeof(struct xt_iprange_mtinfo), .me = THIS_MODULE, diff --git a/net/netfilter/xt_length.c b/net/netfilter/xt_length.c index b8640f97295..c4871ca6c86 100644 --- a/net/netfilter/xt_length.c +++ b/net/netfilter/xt_length.c @@ -21,24 +21,18 @@ MODULE_ALIAS("ipt_length"); MODULE_ALIAS("ip6t_length"); static bool -length_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +length_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_length_info *info = matchinfo; + const struct xt_length_info *info = par->matchinfo; u_int16_t pktlen = ntohs(ip_hdr(skb)->tot_len); return (pktlen >= info->min && pktlen <= info->max) ^ info->invert; } static bool -length_mt6(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +length_mt6(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_length_info *info = matchinfo; + const struct xt_length_info *info = par->matchinfo; const u_int16_t pktlen = ntohs(ipv6_hdr(skb)->payload_len) + sizeof(struct ipv6hdr); @@ -48,14 +42,14 @@ length_mt6(const struct sk_buff *skb, const struct net_device *in, static struct xt_match length_mt_reg[] __read_mostly = { { .name = "length", - .family = AF_INET, + .family = NFPROTO_IPV4, .match = length_mt, .matchsize = sizeof(struct xt_length_info), .me = THIS_MODULE, }, { .name = "length", - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = length_mt6, .matchsize = sizeof(struct xt_length_info), .me = THIS_MODULE, diff --git a/net/netfilter/xt_limit.c b/net/netfilter/xt_limit.c index aad9ab8d204..c908d69a559 100644 --- a/net/netfilter/xt_limit.c +++ b/net/netfilter/xt_limit.c @@ -58,13 +58,10 @@ static DEFINE_SPINLOCK(limit_lock); #define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ) static bool -limit_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +limit_mt(const struct sk_buff *skb, const struct xt_match_param *par) { struct xt_rateinfo *r = - ((const struct xt_rateinfo *)matchinfo)->master; + ((const struct xt_rateinfo *)par->matchinfo)->master; unsigned long now = jiffies; spin_lock_bh(&limit_lock); @@ -95,12 +92,9 @@ user2credits(u_int32_t user) return (user * HZ * CREDITS_PER_JIFFY) / XT_LIMIT_SCALE; } -static bool -limit_mt_check(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool limit_mt_check(const struct xt_mtchk_param *par) { - struct xt_rateinfo *r = matchinfo; + struct xt_rateinfo *r = par->matchinfo; /* Check for overflow. */ if (r->burst == 0 @@ -167,43 +161,29 @@ static int limit_mt_compat_to_user(void __user *dst, void *src) } #endif /* CONFIG_COMPAT */ -static struct xt_match limit_mt_reg[] __read_mostly = { - { - .name = "limit", - .family = AF_INET, - .checkentry = limit_mt_check, - .match = limit_mt, - .matchsize = sizeof(struct xt_rateinfo), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_rateinfo), - .compat_from_user = limit_mt_compat_from_user, - .compat_to_user = limit_mt_compat_to_user, -#endif - .me = THIS_MODULE, - }, - { - .name = "limit", - .family = AF_INET6, - .checkentry = limit_mt_check, - .match = limit_mt, - .matchsize = sizeof(struct xt_rateinfo), +static struct xt_match limit_mt_reg __read_mostly = { + .name = "limit", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = limit_mt, + .checkentry = limit_mt_check, + .matchsize = sizeof(struct xt_rateinfo), #ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_rateinfo), - .compat_from_user = limit_mt_compat_from_user, - .compat_to_user = limit_mt_compat_to_user, + .compatsize = sizeof(struct compat_xt_rateinfo), + .compat_from_user = limit_mt_compat_from_user, + .compat_to_user = limit_mt_compat_to_user, #endif - .me = THIS_MODULE, - }, + .me = THIS_MODULE, }; static int __init limit_mt_init(void) { - return xt_register_matches(limit_mt_reg, ARRAY_SIZE(limit_mt_reg)); + return xt_register_match(&limit_mt_reg); } static void __exit limit_mt_exit(void) { - xt_unregister_matches(limit_mt_reg, ARRAY_SIZE(limit_mt_reg)); + xt_unregister_match(&limit_mt_reg); } module_init(limit_mt_init); diff --git a/net/netfilter/xt_mac.c b/net/netfilter/xt_mac.c index b3e96a0ec17..c2007116ce5 100644 --- a/net/netfilter/xt_mac.c +++ b/net/netfilter/xt_mac.c @@ -24,12 +24,9 @@ MODULE_DESCRIPTION("Xtables: MAC address match"); MODULE_ALIAS("ipt_mac"); MODULE_ALIAS("ip6t_mac"); -static bool -mac_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +static bool mac_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_mac_info *info = matchinfo; + const struct xt_mac_info *info = par->matchinfo; /* Is mac pointer valid? */ return skb_mac_header(skb) >= skb->head && @@ -39,37 +36,25 @@ mac_mt(const struct sk_buff *skb, const struct net_device *in, ^ info->invert); } -static struct xt_match mac_mt_reg[] __read_mostly = { - { - .name = "mac", - .family = AF_INET, - .match = mac_mt, - .matchsize = sizeof(struct xt_mac_info), - .hooks = (1 << NF_INET_PRE_ROUTING) | - (1 << NF_INET_LOCAL_IN) | - (1 << NF_INET_FORWARD), - .me = THIS_MODULE, - }, - { - .name = "mac", - .family = AF_INET6, - .match = mac_mt, - .matchsize = sizeof(struct xt_mac_info), - .hooks = (1 << NF_INET_PRE_ROUTING) | - (1 << NF_INET_LOCAL_IN) | - (1 << NF_INET_FORWARD), - .me = THIS_MODULE, - }, +static struct xt_match mac_mt_reg __read_mostly = { + .name = "mac", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = mac_mt, + .matchsize = sizeof(struct xt_mac_info), + .hooks = (1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN) | + (1 << NF_INET_FORWARD), + .me = THIS_MODULE, }; static int __init mac_mt_init(void) { - return xt_register_matches(mac_mt_reg, ARRAY_SIZE(mac_mt_reg)); + return xt_register_match(&mac_mt_reg); } static void __exit mac_mt_exit(void) { - xt_unregister_matches(mac_mt_reg, ARRAY_SIZE(mac_mt_reg)); + xt_unregister_match(&mac_mt_reg); } module_init(mac_mt_init); diff --git a/net/netfilter/xt_mark.c b/net/netfilter/xt_mark.c index 9f78f6120fb..10b9e34bbc5 100644 --- a/net/netfilter/xt_mark.c +++ b/net/netfilter/xt_mark.c @@ -23,32 +23,24 @@ MODULE_ALIAS("ipt_mark"); MODULE_ALIAS("ip6t_mark"); static bool -mark_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +mark_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_mark_info *info = matchinfo; + const struct xt_mark_info *info = par->matchinfo; return ((skb->mark & info->mask) == info->mark) ^ info->invert; } static bool -mark_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +mark_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_mark_mtinfo1 *info = matchinfo; + const struct xt_mark_mtinfo1 *info = par->matchinfo; return ((skb->mark & info->mask) == info->mark) ^ info->invert; } -static bool -mark_mt_check_v0(const char *tablename, const void *entry, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool mark_mt_check_v0(const struct xt_mtchk_param *par) { - const struct xt_mark_info *minfo = matchinfo; + const struct xt_mark_info *minfo = par->matchinfo; if (minfo->mark > 0xffffffff || minfo->mask > 0xffffffff) { printk(KERN_WARNING "mark: only supports 32bit mark\n"); @@ -92,7 +84,7 @@ static struct xt_match mark_mt_reg[] __read_mostly = { { .name = "mark", .revision = 0, - .family = AF_INET, + .family = NFPROTO_UNSPEC, .checkentry = mark_mt_check_v0, .match = mark_mt_v0, .matchsize = sizeof(struct xt_mark_info), @@ -104,31 +96,9 @@ static struct xt_match mark_mt_reg[] __read_mostly = { .me = THIS_MODULE, }, { - .name = "mark", - .revision = 0, - .family = AF_INET6, - .checkentry = mark_mt_check_v0, - .match = mark_mt_v0, - .matchsize = sizeof(struct xt_mark_info), -#ifdef CONFIG_COMPAT - .compatsize = sizeof(struct compat_xt_mark_info), - .compat_from_user = mark_mt_compat_from_user_v0, - .compat_to_user = mark_mt_compat_to_user_v0, -#endif - .me = THIS_MODULE, - }, - { - .name = "mark", - .revision = 1, - .family = AF_INET, - .match = mark_mt, - .matchsize = sizeof(struct xt_mark_mtinfo1), - .me = THIS_MODULE, - }, - { .name = "mark", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .match = mark_mt, .matchsize = sizeof(struct xt_mark_mtinfo1), .me = THIS_MODULE, diff --git a/net/netfilter/xt_multiport.c b/net/netfilter/xt_multiport.c index fd88c489b70..d06bb2dd390 100644 --- a/net/netfilter/xt_multiport.c +++ b/net/netfilter/xt_multiport.c @@ -95,25 +95,22 @@ ports_match_v1(const struct xt_multiport_v1 *minfo, } static bool -multiport_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +multiport_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { const __be16 *pptr; __be16 _ports[2]; - const struct xt_multiport *multiinfo = matchinfo; + const struct xt_multiport *multiinfo = par->matchinfo; - if (offset) + if (par->fragoff != 0) return false; - pptr = skb_header_pointer(skb, protoff, sizeof(_ports), _ports); + pptr = skb_header_pointer(skb, par->thoff, sizeof(_ports), _ports); if (pptr == NULL) { /* We've been asked to examine this packet, and we * can't. Hence, no choice but to drop. */ duprintf("xt_multiport: Dropping evil offset=0 tinygram.\n"); - *hotdrop = true; + *par->hotdrop = true; return false; } @@ -122,25 +119,22 @@ multiport_mt_v0(const struct sk_buff *skb, const struct net_device *in, } static bool -multiport_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +multiport_mt(const struct sk_buff *skb, const struct xt_match_param *par) { const __be16 *pptr; __be16 _ports[2]; - const struct xt_multiport_v1 *multiinfo = matchinfo; + const struct xt_multiport_v1 *multiinfo = par->matchinfo; - if (offset) + if (par->fragoff != 0) return false; - pptr = skb_header_pointer(skb, protoff, sizeof(_ports), _ports); + pptr = skb_header_pointer(skb, par->thoff, sizeof(_ports), _ports); if (pptr == NULL) { /* We've been asked to examine this packet, and we * can't. Hence, no choice but to drop. */ duprintf("xt_multiport: Dropping evil offset=0 tinygram.\n"); - *hotdrop = true; + *par->hotdrop = true; return false; } @@ -164,50 +158,37 @@ check(u_int16_t proto, && count <= XT_MULTI_PORTS; } -/* Called when user tries to insert an entry of this type. */ -static bool -multiport_mt_check_v0(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool multiport_mt_check_v0(const struct xt_mtchk_param *par) { - const struct ipt_ip *ip = info; - const struct xt_multiport *multiinfo = matchinfo; + const struct ipt_ip *ip = par->entryinfo; + const struct xt_multiport *multiinfo = par->matchinfo; return check(ip->proto, ip->invflags, multiinfo->flags, multiinfo->count); } -static bool -multiport_mt_check(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool multiport_mt_check(const struct xt_mtchk_param *par) { - const struct ipt_ip *ip = info; - const struct xt_multiport_v1 *multiinfo = matchinfo; + const struct ipt_ip *ip = par->entryinfo; + const struct xt_multiport_v1 *multiinfo = par->matchinfo; return check(ip->proto, ip->invflags, multiinfo->flags, multiinfo->count); } -static bool -multiport_mt6_check_v0(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool multiport_mt6_check_v0(const struct xt_mtchk_param *par) { - const struct ip6t_ip6 *ip = info; - const struct xt_multiport *multiinfo = matchinfo; + const struct ip6t_ip6 *ip = par->entryinfo; + const struct xt_multiport *multiinfo = par->matchinfo; return check(ip->proto, ip->invflags, multiinfo->flags, multiinfo->count); } -static bool -multiport_mt6_check(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool multiport_mt6_check(const struct xt_mtchk_param *par) { - const struct ip6t_ip6 *ip = info; - const struct xt_multiport_v1 *multiinfo = matchinfo; + const struct ip6t_ip6 *ip = par->entryinfo; + const struct xt_multiport_v1 *multiinfo = par->matchinfo; return check(ip->proto, ip->invflags, multiinfo->flags, multiinfo->count); @@ -216,7 +197,7 @@ multiport_mt6_check(const char *tablename, const void *info, static struct xt_match multiport_mt_reg[] __read_mostly = { { .name = "multiport", - .family = AF_INET, + .family = NFPROTO_IPV4, .revision = 0, .checkentry = multiport_mt_check_v0, .match = multiport_mt_v0, @@ -225,7 +206,7 @@ static struct xt_match multiport_mt_reg[] __read_mostly = { }, { .name = "multiport", - .family = AF_INET, + .family = NFPROTO_IPV4, .revision = 1, .checkentry = multiport_mt_check, .match = multiport_mt, @@ -234,7 +215,7 @@ static struct xt_match multiport_mt_reg[] __read_mostly = { }, { .name = "multiport", - .family = AF_INET6, + .family = NFPROTO_IPV6, .revision = 0, .checkentry = multiport_mt6_check_v0, .match = multiport_mt_v0, @@ -243,7 +224,7 @@ static struct xt_match multiport_mt_reg[] __read_mostly = { }, { .name = "multiport", - .family = AF_INET6, + .family = NFPROTO_IPV6, .revision = 1, .checkentry = multiport_mt6_check, .match = multiport_mt, diff --git a/net/netfilter/xt_owner.c b/net/netfilter/xt_owner.c index 9059c16144c..f19ebd9b78f 100644 --- a/net/netfilter/xt_owner.c +++ b/net/netfilter/xt_owner.c @@ -21,12 +21,9 @@ #include <linux/netfilter_ipv6/ip6t_owner.h> static bool -owner_mt_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +owner_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct ipt_owner_info *info = matchinfo; + const struct ipt_owner_info *info = par->matchinfo; const struct file *filp; if (skb->sk == NULL || skb->sk->sk_socket == NULL) @@ -50,12 +47,9 @@ owner_mt_v0(const struct sk_buff *skb, const struct net_device *in, } static bool -owner_mt6_v0(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +owner_mt6_v0(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct ip6t_owner_info *info = matchinfo; + const struct ip6t_owner_info *info = par->matchinfo; const struct file *filp; if (skb->sk == NULL || skb->sk->sk_socket == NULL) @@ -79,12 +73,9 @@ owner_mt6_v0(const struct sk_buff *skb, const struct net_device *in, } static bool -owner_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +owner_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_owner_match_info *info = matchinfo; + const struct xt_owner_match_info *info = par->matchinfo; const struct file *filp; if (skb->sk == NULL || skb->sk->sk_socket == NULL) @@ -116,12 +107,9 @@ owner_mt(const struct sk_buff *skb, const struct net_device *in, return true; } -static bool -owner_mt_check_v0(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool owner_mt_check_v0(const struct xt_mtchk_param *par) { - const struct ipt_owner_info *info = matchinfo; + const struct ipt_owner_info *info = par->matchinfo; if (info->match & (IPT_OWNER_PID | IPT_OWNER_SID | IPT_OWNER_COMM)) { printk(KERN_WARNING KBUILD_MODNAME @@ -133,12 +121,9 @@ owner_mt_check_v0(const char *tablename, const void *ip, return true; } -static bool -owner_mt6_check_v0(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool owner_mt6_check_v0(const struct xt_mtchk_param *par) { - const struct ip6t_owner_info *info = matchinfo; + const struct ip6t_owner_info *info = par->matchinfo; if (info->match & (IP6T_OWNER_PID | IP6T_OWNER_SID)) { printk(KERN_WARNING KBUILD_MODNAME @@ -153,7 +138,7 @@ static struct xt_match owner_mt_reg[] __read_mostly = { { .name = "owner", .revision = 0, - .family = AF_INET, + .family = NFPROTO_IPV4, .match = owner_mt_v0, .matchsize = sizeof(struct ipt_owner_info), .checkentry = owner_mt_check_v0, @@ -164,7 +149,7 @@ static struct xt_match owner_mt_reg[] __read_mostly = { { .name = "owner", .revision = 0, - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = owner_mt6_v0, .matchsize = sizeof(struct ip6t_owner_info), .checkentry = owner_mt6_check_v0, @@ -175,17 +160,7 @@ static struct xt_match owner_mt_reg[] __read_mostly = { { .name = "owner", .revision = 1, - .family = AF_INET, - .match = owner_mt, - .matchsize = sizeof(struct xt_owner_match_info), - .hooks = (1 << NF_INET_LOCAL_OUT) | - (1 << NF_INET_POST_ROUTING), - .me = THIS_MODULE, - }, - { - .name = "owner", - .revision = 1, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .match = owner_mt, .matchsize = sizeof(struct xt_owner_match_info), .hooks = (1 << NF_INET_LOCAL_OUT) | diff --git a/net/netfilter/xt_physdev.c b/net/netfilter/xt_physdev.c index 4ec1094bda9..1bcdfc12cf5 100644 --- a/net/netfilter/xt_physdev.c +++ b/net/netfilter/xt_physdev.c @@ -21,14 +21,11 @@ MODULE_ALIAS("ipt_physdev"); MODULE_ALIAS("ip6t_physdev"); static bool -physdev_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +physdev_mt(const struct sk_buff *skb, const struct xt_match_param *par) { int i; static const char nulldevname[IFNAMSIZ]; - const struct xt_physdev_info *info = matchinfo; + const struct xt_physdev_info *info = par->matchinfo; bool ret; const char *indev, *outdev; const struct nf_bridge_info *nf_bridge; @@ -94,12 +91,9 @@ match_outdev: return ret ^ !(info->invert & XT_PHYSDEV_OP_OUT); } -static bool -physdev_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool physdev_mt_check(const struct xt_mtchk_param *par) { - const struct xt_physdev_info *info = matchinfo; + const struct xt_physdev_info *info = par->matchinfo; if (!(info->bitmask & XT_PHYSDEV_OP_MASK) || info->bitmask & ~XT_PHYSDEV_OP_MASK) @@ -107,44 +101,35 @@ physdev_mt_check(const char *tablename, const void *ip, if (info->bitmask & XT_PHYSDEV_OP_OUT && (!(info->bitmask & XT_PHYSDEV_OP_BRIDGED) || info->invert & XT_PHYSDEV_OP_BRIDGED) && - hook_mask & ((1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_FORWARD) | - (1 << NF_INET_POST_ROUTING))) { + par->hook_mask & ((1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_FORWARD) | (1 << NF_INET_POST_ROUTING))) { printk(KERN_WARNING "physdev match: using --physdev-out in the " "OUTPUT, FORWARD and POSTROUTING chains for non-bridged " "traffic is not supported anymore.\n"); - if (hook_mask & (1 << NF_INET_LOCAL_OUT)) + if (par->hook_mask & (1 << NF_INET_LOCAL_OUT)) return false; } return true; } -static struct xt_match physdev_mt_reg[] __read_mostly = { - { - .name = "physdev", - .family = AF_INET, - .checkentry = physdev_mt_check, - .match = physdev_mt, - .matchsize = sizeof(struct xt_physdev_info), - .me = THIS_MODULE, - }, - { - .name = "physdev", - .family = AF_INET6, - .checkentry = physdev_mt_check, - .match = physdev_mt, - .matchsize = sizeof(struct xt_physdev_info), - .me = THIS_MODULE, - }, +static struct xt_match physdev_mt_reg __read_mostly = { + .name = "physdev", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = physdev_mt_check, + .match = physdev_mt, + .matchsize = sizeof(struct xt_physdev_info), + .me = THIS_MODULE, }; static int __init physdev_mt_init(void) { - return xt_register_matches(physdev_mt_reg, ARRAY_SIZE(physdev_mt_reg)); + return xt_register_match(&physdev_mt_reg); } static void __exit physdev_mt_exit(void) { - xt_unregister_matches(physdev_mt_reg, ARRAY_SIZE(physdev_mt_reg)); + xt_unregister_match(&physdev_mt_reg); } module_init(physdev_mt_init); diff --git a/net/netfilter/xt_pkttype.c b/net/netfilter/xt_pkttype.c index 7936f7e2325..69da1d3a1d8 100644 --- a/net/netfilter/xt_pkttype.c +++ b/net/netfilter/xt_pkttype.c @@ -23,20 +23,17 @@ MODULE_ALIAS("ipt_pkttype"); MODULE_ALIAS("ip6t_pkttype"); static bool -pkttype_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +pkttype_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_pkttype_info *info = matchinfo; + const struct xt_pkttype_info *info = par->matchinfo; u_int8_t type; if (skb->pkt_type != PACKET_LOOPBACK) type = skb->pkt_type; - else if (match->family == AF_INET && + else if (par->family == NFPROTO_IPV4 && ipv4_is_multicast(ip_hdr(skb)->daddr)) type = PACKET_MULTICAST; - else if (match->family == AF_INET6 && + else if (par->family == NFPROTO_IPV6 && ipv6_hdr(skb)->daddr.s6_addr[0] == 0xFF) type = PACKET_MULTICAST; else @@ -45,31 +42,23 @@ pkttype_mt(const struct sk_buff *skb, const struct net_device *in, return (type == info->pkttype) ^ info->invert; } -static struct xt_match pkttype_mt_reg[] __read_mostly = { - { - .name = "pkttype", - .family = AF_INET, - .match = pkttype_mt, - .matchsize = sizeof(struct xt_pkttype_info), - .me = THIS_MODULE, - }, - { - .name = "pkttype", - .family = AF_INET6, - .match = pkttype_mt, - .matchsize = sizeof(struct xt_pkttype_info), - .me = THIS_MODULE, - }, +static struct xt_match pkttype_mt_reg __read_mostly = { + .name = "pkttype", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = pkttype_mt, + .matchsize = sizeof(struct xt_pkttype_info), + .me = THIS_MODULE, }; static int __init pkttype_mt_init(void) { - return xt_register_matches(pkttype_mt_reg, ARRAY_SIZE(pkttype_mt_reg)); + return xt_register_match(&pkttype_mt_reg); } static void __exit pkttype_mt_exit(void) { - xt_unregister_matches(pkttype_mt_reg, ARRAY_SIZE(pkttype_mt_reg)); + xt_unregister_match(&pkttype_mt_reg); } module_init(pkttype_mt_init); diff --git a/net/netfilter/xt_policy.c b/net/netfilter/xt_policy.c index d351582b2a3..328bd20ddd2 100644 --- a/net/netfilter/xt_policy.c +++ b/net/netfilter/xt_policy.c @@ -26,9 +26,9 @@ xt_addr_cmp(const union nf_inet_addr *a1, const union nf_inet_addr *m, const union nf_inet_addr *a2, unsigned short family) { switch (family) { - case AF_INET: + case NFPROTO_IPV4: return ((a1->ip ^ a2->ip) & m->ip) == 0; - case AF_INET6: + case NFPROTO_IPV6: return ipv6_masked_addr_cmp(&a1->in6, &m->in6, &a2->in6) == 0; } return false; @@ -110,18 +110,15 @@ match_policy_out(const struct sk_buff *skb, const struct xt_policy_info *info, } static bool -policy_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +policy_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_policy_info *info = matchinfo; + const struct xt_policy_info *info = par->matchinfo; int ret; if (info->flags & XT_POLICY_MATCH_IN) - ret = match_policy_in(skb, info, match->family); + ret = match_policy_in(skb, info, par->match->family); else - ret = match_policy_out(skb, info, match->family); + ret = match_policy_out(skb, info, par->match->family); if (ret < 0) ret = info->flags & XT_POLICY_MATCH_NONE ? true : false; @@ -131,26 +128,23 @@ policy_mt(const struct sk_buff *skb, const struct net_device *in, return ret; } -static bool -policy_mt_check(const char *tablename, const void *ip_void, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool policy_mt_check(const struct xt_mtchk_param *par) { - const struct xt_policy_info *info = matchinfo; + const struct xt_policy_info *info = par->matchinfo; if (!(info->flags & (XT_POLICY_MATCH_IN|XT_POLICY_MATCH_OUT))) { printk(KERN_ERR "xt_policy: neither incoming nor " "outgoing policy selected\n"); return false; } - if (hook_mask & (1 << NF_INET_PRE_ROUTING | 1 << NF_INET_LOCAL_IN) - && info->flags & XT_POLICY_MATCH_OUT) { + if (par->hook_mask & ((1 << NF_INET_PRE_ROUTING) | + (1 << NF_INET_LOCAL_IN)) && info->flags & XT_POLICY_MATCH_OUT) { printk(KERN_ERR "xt_policy: output policy not valid in " "PRE_ROUTING and INPUT\n"); return false; } - if (hook_mask & (1 << NF_INET_POST_ROUTING | 1 << NF_INET_LOCAL_OUT) - && info->flags & XT_POLICY_MATCH_IN) { + if (par->hook_mask & ((1 << NF_INET_POST_ROUTING) | + (1 << NF_INET_LOCAL_OUT)) && info->flags & XT_POLICY_MATCH_IN) { printk(KERN_ERR "xt_policy: input policy not valid in " "POST_ROUTING and OUTPUT\n"); return false; @@ -165,7 +159,7 @@ policy_mt_check(const char *tablename, const void *ip_void, static struct xt_match policy_mt_reg[] __read_mostly = { { .name = "policy", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = policy_mt_check, .match = policy_mt, .matchsize = sizeof(struct xt_policy_info), @@ -173,7 +167,7 @@ static struct xt_match policy_mt_reg[] __read_mostly = { }, { .name = "policy", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = policy_mt_check, .match = policy_mt, .matchsize = sizeof(struct xt_policy_info), diff --git a/net/netfilter/xt_quota.c b/net/netfilter/xt_quota.c index 3b021d0c522..c84fce5e0f3 100644 --- a/net/netfilter/xt_quota.c +++ b/net/netfilter/xt_quota.c @@ -18,13 +18,10 @@ MODULE_ALIAS("ip6t_quota"); static DEFINE_SPINLOCK(quota_lock); static bool -quota_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +quota_mt(const struct sk_buff *skb, const struct xt_match_param *par) { struct xt_quota_info *q = - ((const struct xt_quota_info *)matchinfo)->master; + ((const struct xt_quota_info *)par->matchinfo)->master; bool ret = q->flags & XT_QUOTA_INVERT; spin_lock_bh("a_lock); @@ -40,12 +37,9 @@ quota_mt(const struct sk_buff *skb, const struct net_device *in, return ret; } -static bool -quota_mt_check(const char *tablename, const void *entry, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool quota_mt_check(const struct xt_mtchk_param *par) { - struct xt_quota_info *q = matchinfo; + struct xt_quota_info *q = par->matchinfo; if (q->flags & ~XT_QUOTA_MASK) return false; @@ -54,33 +48,24 @@ quota_mt_check(const char *tablename, const void *entry, return true; } -static struct xt_match quota_mt_reg[] __read_mostly = { - { - .name = "quota", - .family = AF_INET, - .checkentry = quota_mt_check, - .match = quota_mt, - .matchsize = sizeof(struct xt_quota_info), - .me = THIS_MODULE - }, - { - .name = "quota", - .family = AF_INET6, - .checkentry = quota_mt_check, - .match = quota_mt, - .matchsize = sizeof(struct xt_quota_info), - .me = THIS_MODULE - }, +static struct xt_match quota_mt_reg __read_mostly = { + .name = "quota", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = quota_mt, + .checkentry = quota_mt_check, + .matchsize = sizeof(struct xt_quota_info), + .me = THIS_MODULE, }; static int __init quota_mt_init(void) { - return xt_register_matches(quota_mt_reg, ARRAY_SIZE(quota_mt_reg)); + return xt_register_match("a_mt_reg); } static void __exit quota_mt_exit(void) { - xt_unregister_matches(quota_mt_reg, ARRAY_SIZE(quota_mt_reg)); + xt_unregister_match("a_mt_reg); } module_init(quota_mt_init); diff --git a/net/netfilter/xt_rateest.c b/net/netfilter/xt_rateest.c index ebd84f1b4f6..220a1d588ee 100644 --- a/net/netfilter/xt_rateest.c +++ b/net/netfilter/xt_rateest.c @@ -14,16 +14,10 @@ #include <net/netfilter/xt_rateest.h> -static bool xt_rateest_mt(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const struct xt_match *match, - const void *matchinfo, - int offset, - unsigned int protoff, - bool *hotdrop) +static bool +xt_rateest_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_rateest_match_info *info = matchinfo; + const struct xt_rateest_match_info *info = par->matchinfo; struct gnet_stats_rate_est *r; u_int32_t bps1, bps2, pps1, pps2; bool ret = true; @@ -80,13 +74,9 @@ static bool xt_rateest_mt(const struct sk_buff *skb, return ret; } -static bool xt_rateest_mt_checkentry(const char *tablename, - const void *ip, - const struct xt_match *match, - void *matchinfo, - unsigned int hook_mask) +static bool xt_rateest_mt_checkentry(const struct xt_mtchk_param *par) { - struct xt_rateest_match_info *info = matchinfo; + struct xt_rateest_match_info *info = par->matchinfo; struct xt_rateest *est1, *est2; if (hweight32(info->flags & (XT_RATEEST_MATCH_ABS | @@ -127,46 +117,34 @@ err1: return false; } -static void xt_rateest_mt_destroy(const struct xt_match *match, - void *matchinfo) +static void xt_rateest_mt_destroy(const struct xt_mtdtor_param *par) { - struct xt_rateest_match_info *info = matchinfo; + struct xt_rateest_match_info *info = par->matchinfo; xt_rateest_put(info->est1); if (info->est2) xt_rateest_put(info->est2); } -static struct xt_match xt_rateest_match[] __read_mostly = { - { - .family = AF_INET, - .name = "rateest", - .match = xt_rateest_mt, - .checkentry = xt_rateest_mt_checkentry, - .destroy = xt_rateest_mt_destroy, - .matchsize = sizeof(struct xt_rateest_match_info), - .me = THIS_MODULE, - }, - { - .family = AF_INET6, - .name = "rateest", - .match = xt_rateest_mt, - .checkentry = xt_rateest_mt_checkentry, - .destroy = xt_rateest_mt_destroy, - .matchsize = sizeof(struct xt_rateest_match_info), - .me = THIS_MODULE, - }, +static struct xt_match xt_rateest_mt_reg __read_mostly = { + .name = "rateest", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = xt_rateest_mt, + .checkentry = xt_rateest_mt_checkentry, + .destroy = xt_rateest_mt_destroy, + .matchsize = sizeof(struct xt_rateest_match_info), + .me = THIS_MODULE, }; static int __init xt_rateest_mt_init(void) { - return xt_register_matches(xt_rateest_match, - ARRAY_SIZE(xt_rateest_match)); + return xt_register_match(&xt_rateest_mt_reg); } static void __exit xt_rateest_mt_fini(void) { - xt_unregister_matches(xt_rateest_match, ARRAY_SIZE(xt_rateest_match)); + xt_unregister_match(&xt_rateest_mt_reg); } MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); diff --git a/net/netfilter/xt_realm.c b/net/netfilter/xt_realm.c index 7df1627c536..67419287bc7 100644 --- a/net/netfilter/xt_realm.c +++ b/net/netfilter/xt_realm.c @@ -22,12 +22,9 @@ MODULE_DESCRIPTION("Xtables: Routing realm match"); MODULE_ALIAS("ipt_realm"); static bool -realm_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +realm_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_realm_info *info = matchinfo; + const struct xt_realm_info *info = par->matchinfo; const struct dst_entry *dst = skb->dst; return (info->id == (dst->tclassid & info->mask)) ^ info->invert; @@ -39,7 +36,7 @@ static struct xt_match realm_mt_reg __read_mostly = { .matchsize = sizeof(struct xt_realm_info), .hooks = (1 << NF_INET_POST_ROUTING) | (1 << NF_INET_FORWARD) | (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_LOCAL_IN), - .family = AF_INET, + .family = NFPROTO_UNSPEC, .me = THIS_MODULE }; diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c new file mode 100644 index 00000000000..4ebd4ca9a99 --- /dev/null +++ b/net/netfilter/xt_recent.c @@ -0,0 +1,689 @@ +/* + * Copyright (c) 2006 Patrick McHardy <kaber@trash.net> + * Copyright © CC Computer Consultants GmbH, 2007 - 2008 + * + * 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. + * + * This is a replacement of the old ipt_recent module, which carried the + * following copyright notice: + * + * Author: Stephen Frost <sfrost@snowman.net> + * Copyright 2002-2003, Stephen Frost, 2.5.x port by laforge@netfilter.org + */ +#include <linux/init.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/string.h> +#include <linux/ctype.h> +#include <linux/list.h> +#include <linux/random.h> +#include <linux/jhash.h> +#include <linux/bitops.h> +#include <linux/skbuff.h> +#include <linux/inet.h> +#include <net/net_namespace.h> + +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_recent.h> + +MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); +MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>"); +MODULE_DESCRIPTION("Xtables: \"recently-seen\" host matching for IPv4"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_recent"); +MODULE_ALIAS("ip6t_recent"); + +static unsigned int ip_list_tot = 100; +static unsigned int ip_pkt_list_tot = 20; +static unsigned int ip_list_hash_size = 0; +static unsigned int ip_list_perms = 0644; +static unsigned int ip_list_uid = 0; +static unsigned int ip_list_gid = 0; +module_param(ip_list_tot, uint, 0400); +module_param(ip_pkt_list_tot, uint, 0400); +module_param(ip_list_hash_size, uint, 0400); +module_param(ip_list_perms, uint, 0400); +module_param(ip_list_uid, uint, 0400); +module_param(ip_list_gid, uint, 0400); +MODULE_PARM_DESC(ip_list_tot, "number of IPs to remember per list"); +MODULE_PARM_DESC(ip_pkt_list_tot, "number of packets per IP to remember (max. 255)"); +MODULE_PARM_DESC(ip_list_hash_size, "size of hash table used to look up IPs"); +MODULE_PARM_DESC(ip_list_perms, "permissions on /proc/net/xt_recent/* files"); +MODULE_PARM_DESC(ip_list_uid,"owner of /proc/net/xt_recent/* files"); +MODULE_PARM_DESC(ip_list_gid,"owning group of /proc/net/xt_recent/* files"); + +struct recent_entry { + struct list_head list; + struct list_head lru_list; + union nf_inet_addr addr; + u_int16_t family; + u_int8_t ttl; + u_int8_t index; + u_int16_t nstamps; + unsigned long stamps[0]; +}; + +struct recent_table { + struct list_head list; + char name[XT_RECENT_NAME_LEN]; +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *proc_old, *proc; +#endif + unsigned int refcnt; + unsigned int entries; + struct list_head lru_list; + struct list_head iphash[0]; +}; + +static LIST_HEAD(tables); +static DEFINE_SPINLOCK(recent_lock); +static DEFINE_MUTEX(recent_mutex); + +#ifdef CONFIG_PROC_FS +#ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT +static struct proc_dir_entry *proc_old_dir; +#endif +static struct proc_dir_entry *recent_proc_dir; +static const struct file_operations recent_old_fops, recent_mt_fops; +#endif + +static u_int32_t hash_rnd; +static bool hash_rnd_initted; + +static unsigned int recent_entry_hash4(const union nf_inet_addr *addr) +{ + if (!hash_rnd_initted) { + get_random_bytes(&hash_rnd, sizeof(hash_rnd)); + hash_rnd_initted = true; + } + return jhash_1word((__force u32)addr->ip, hash_rnd) & + (ip_list_hash_size - 1); +} + +static unsigned int recent_entry_hash6(const union nf_inet_addr *addr) +{ + if (!hash_rnd_initted) { + get_random_bytes(&hash_rnd, sizeof(hash_rnd)); + hash_rnd_initted = true; + } + return jhash2((u32 *)addr->ip6, ARRAY_SIZE(addr->ip6), hash_rnd) & + (ip_list_hash_size - 1); +} + +static struct recent_entry * +recent_entry_lookup(const struct recent_table *table, + const union nf_inet_addr *addrp, u_int16_t family, + u_int8_t ttl) +{ + struct recent_entry *e; + unsigned int h; + + if (family == NFPROTO_IPV4) + h = recent_entry_hash4(addrp); + else + h = recent_entry_hash6(addrp); + + list_for_each_entry(e, &table->iphash[h], list) + if (e->family == family && + memcmp(&e->addr, addrp, sizeof(e->addr)) == 0 && + (ttl == e->ttl || ttl == 0 || e->ttl == 0)) + return e; + return NULL; +} + +static void recent_entry_remove(struct recent_table *t, struct recent_entry *e) +{ + list_del(&e->list); + list_del(&e->lru_list); + kfree(e); + t->entries--; +} + +static struct recent_entry * +recent_entry_init(struct recent_table *t, const union nf_inet_addr *addr, + u_int16_t family, u_int8_t ttl) +{ + struct recent_entry *e; + + if (t->entries >= ip_list_tot) { + e = list_entry(t->lru_list.next, struct recent_entry, lru_list); + recent_entry_remove(t, e); + } + e = kmalloc(sizeof(*e) + sizeof(e->stamps[0]) * ip_pkt_list_tot, + GFP_ATOMIC); + if (e == NULL) + return NULL; + memcpy(&e->addr, addr, sizeof(e->addr)); + e->ttl = ttl; + e->stamps[0] = jiffies; + e->nstamps = 1; + e->index = 1; + e->family = family; + if (family == NFPROTO_IPV4) + list_add_tail(&e->list, &t->iphash[recent_entry_hash4(addr)]); + else + list_add_tail(&e->list, &t->iphash[recent_entry_hash6(addr)]); + list_add_tail(&e->lru_list, &t->lru_list); + t->entries++; + return e; +} + +static void recent_entry_update(struct recent_table *t, struct recent_entry *e) +{ + e->stamps[e->index++] = jiffies; + if (e->index > e->nstamps) + e->nstamps = e->index; + e->index %= ip_pkt_list_tot; + list_move_tail(&e->lru_list, &t->lru_list); +} + +static struct recent_table *recent_table_lookup(const char *name) +{ + struct recent_table *t; + + list_for_each_entry(t, &tables, list) + if (!strcmp(t->name, name)) + return t; + return NULL; +} + +static void recent_table_flush(struct recent_table *t) +{ + struct recent_entry *e, *next; + unsigned int i; + + for (i = 0; i < ip_list_hash_size; i++) + list_for_each_entry_safe(e, next, &t->iphash[i], list) + recent_entry_remove(t, e); +} + +static bool +recent_mt(const struct sk_buff *skb, const struct xt_match_param *par) +{ + const struct xt_recent_mtinfo *info = par->matchinfo; + struct recent_table *t; + struct recent_entry *e; + union nf_inet_addr addr = {}; + u_int8_t ttl; + bool ret = info->invert; + + if (par->match->family == NFPROTO_IPV4) { + const struct iphdr *iph = ip_hdr(skb); + + if (info->side == XT_RECENT_DEST) + addr.ip = iph->daddr; + else + addr.ip = iph->saddr; + + ttl = iph->ttl; + } else { + const struct ipv6hdr *iph = ipv6_hdr(skb); + + if (info->side == XT_RECENT_DEST) + memcpy(&addr.in6, &iph->daddr, sizeof(addr.in6)); + else + memcpy(&addr.in6, &iph->saddr, sizeof(addr.in6)); + + ttl = iph->hop_limit; + } + + /* use TTL as seen before forwarding */ + if (par->out != NULL && skb->sk == NULL) + ttl++; + + spin_lock_bh(&recent_lock); + t = recent_table_lookup(info->name); + e = recent_entry_lookup(t, &addr, par->match->family, + (info->check_set & XT_RECENT_TTL) ? ttl : 0); + if (e == NULL) { + if (!(info->check_set & XT_RECENT_SET)) + goto out; + e = recent_entry_init(t, &addr, par->match->family, ttl); + if (e == NULL) + *par->hotdrop = true; + ret = !ret; + goto out; + } + + if (info->check_set & XT_RECENT_SET) + ret = !ret; + else if (info->check_set & XT_RECENT_REMOVE) { + recent_entry_remove(t, e); + ret = !ret; + } else if (info->check_set & (XT_RECENT_CHECK | XT_RECENT_UPDATE)) { + unsigned long time = jiffies - info->seconds * HZ; + unsigned int i, hits = 0; + + for (i = 0; i < e->nstamps; i++) { + if (info->seconds && time_after(time, e->stamps[i])) + continue; + if (++hits >= info->hit_count) { + ret = !ret; + break; + } + } + } + + if (info->check_set & XT_RECENT_SET || + (info->check_set & XT_RECENT_UPDATE && ret)) { + recent_entry_update(t, e); + e->ttl = ttl; + } +out: + spin_unlock_bh(&recent_lock); + return ret; +} + +static bool recent_mt_check(const struct xt_mtchk_param *par) +{ + const struct xt_recent_mtinfo *info = par->matchinfo; + struct recent_table *t; + unsigned i; + bool ret = false; + + if (hweight8(info->check_set & + (XT_RECENT_SET | XT_RECENT_REMOVE | + XT_RECENT_CHECK | XT_RECENT_UPDATE)) != 1) + return false; + if ((info->check_set & (XT_RECENT_SET | XT_RECENT_REMOVE)) && + (info->seconds || info->hit_count)) + return false; + if (info->hit_count > ip_pkt_list_tot) + return false; + if (info->name[0] == '\0' || + strnlen(info->name, XT_RECENT_NAME_LEN) == XT_RECENT_NAME_LEN) + return false; + + mutex_lock(&recent_mutex); + t = recent_table_lookup(info->name); + if (t != NULL) { + t->refcnt++; + ret = true; + goto out; + } + + t = kzalloc(sizeof(*t) + sizeof(t->iphash[0]) * ip_list_hash_size, + GFP_KERNEL); + if (t == NULL) + goto out; + t->refcnt = 1; + strcpy(t->name, info->name); + INIT_LIST_HEAD(&t->lru_list); + for (i = 0; i < ip_list_hash_size; i++) + INIT_LIST_HEAD(&t->iphash[i]); +#ifdef CONFIG_PROC_FS + t->proc = proc_create(t->name, ip_list_perms, recent_proc_dir, + &recent_mt_fops); + if (t->proc == NULL) { + kfree(t); + goto out; + } +#ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT + t->proc_old = proc_create(t->name, ip_list_perms, proc_old_dir, + &recent_old_fops); + if (t->proc_old == NULL) { + remove_proc_entry(t->name, proc_old_dir); + kfree(t); + goto out; + } + t->proc_old->uid = ip_list_uid; + t->proc_old->gid = ip_list_gid; + t->proc_old->data = t; +#endif + t->proc->uid = ip_list_uid; + t->proc->gid = ip_list_gid; + t->proc->data = t; +#endif + spin_lock_bh(&recent_lock); + list_add_tail(&t->list, &tables); + spin_unlock_bh(&recent_lock); + ret = true; +out: + mutex_unlock(&recent_mutex); + return ret; +} + +static void recent_mt_destroy(const struct xt_mtdtor_param *par) +{ + const struct xt_recent_mtinfo *info = par->matchinfo; + struct recent_table *t; + + mutex_lock(&recent_mutex); + t = recent_table_lookup(info->name); + if (--t->refcnt == 0) { + spin_lock_bh(&recent_lock); + list_del(&t->list); + spin_unlock_bh(&recent_lock); +#ifdef CONFIG_PROC_FS +#ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT + remove_proc_entry(t->name, proc_old_dir); +#endif + remove_proc_entry(t->name, recent_proc_dir); +#endif + recent_table_flush(t); + kfree(t); + } + mutex_unlock(&recent_mutex); +} + +#ifdef CONFIG_PROC_FS +struct recent_iter_state { + const struct recent_table *table; + unsigned int bucket; +}; + +static void *recent_seq_start(struct seq_file *seq, loff_t *pos) + __acquires(recent_lock) +{ + struct recent_iter_state *st = seq->private; + const struct recent_table *t = st->table; + struct recent_entry *e; + loff_t p = *pos; + + spin_lock_bh(&recent_lock); + + for (st->bucket = 0; st->bucket < ip_list_hash_size; st->bucket++) + list_for_each_entry(e, &t->iphash[st->bucket], list) + if (p-- == 0) + return e; + return NULL; +} + +static void *recent_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct recent_iter_state *st = seq->private; + const struct recent_table *t = st->table; + const struct recent_entry *e = v; + const struct list_head *head = e->list.next; + + while (head == &t->iphash[st->bucket]) { + if (++st->bucket >= ip_list_hash_size) + return NULL; + head = t->iphash[st->bucket].next; + } + (*pos)++; + return list_entry(head, struct recent_entry, list); +} + +static void recent_seq_stop(struct seq_file *s, void *v) + __releases(recent_lock) +{ + spin_unlock_bh(&recent_lock); +} + +static int recent_seq_show(struct seq_file *seq, void *v) +{ + const struct recent_entry *e = v; + unsigned int i; + + i = (e->index - 1) % ip_pkt_list_tot; + if (e->family == NFPROTO_IPV4) + seq_printf(seq, "src=" NIPQUAD_FMT " ttl: %u last_seen: %lu " + "oldest_pkt: %u", NIPQUAD(e->addr.ip), e->ttl, + e->stamps[i], e->index); + else + seq_printf(seq, "src=" NIP6_FMT " ttl: %u last_seen: %lu " + "oldest_pkt: %u", NIP6(e->addr.in6), e->ttl, + e->stamps[i], e->index); + for (i = 0; i < e->nstamps; i++) + seq_printf(seq, "%s %lu", i ? "," : "", e->stamps[i]); + seq_printf(seq, "\n"); + return 0; +} + +static const struct seq_operations recent_seq_ops = { + .start = recent_seq_start, + .next = recent_seq_next, + .stop = recent_seq_stop, + .show = recent_seq_show, +}; + +static int recent_seq_open(struct inode *inode, struct file *file) +{ + struct proc_dir_entry *pde = PDE(inode); + struct recent_iter_state *st; + + st = __seq_open_private(file, &recent_seq_ops, sizeof(*st)); + if (st == NULL) + return -ENOMEM; + + st->table = pde->data; + return 0; +} + +#ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT +static int recent_old_seq_open(struct inode *inode, struct file *filp) +{ + static bool warned_of_old; + + if (unlikely(!warned_of_old)) { + printk(KERN_INFO KBUILD_MODNAME ": Use of /proc/net/ipt_recent" + " is deprecated; use /proc/net/xt_recent.\n"); + warned_of_old = true; + } + return recent_seq_open(inode, filp); +} + +static ssize_t recent_old_proc_write(struct file *file, + const char __user *input, + size_t size, loff_t *loff) +{ + const struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode); + struct recent_table *t = pde->data; + struct recent_entry *e; + char buf[sizeof("+255.255.255.255")], *c = buf; + __be32 addr; + int add; + + if (size > sizeof(buf)) + size = sizeof(buf); + if (copy_from_user(buf, input, size)) + return -EFAULT; + + while (isspace(*c)) + c++; + + if (size - (c - buf) < 5) + return c - buf; + if (!strncmp(c, "clear", 5)) { + c += 5; + spin_lock_bh(&recent_lock); + recent_table_flush(t); + spin_unlock_bh(&recent_lock); + return c - buf; + } + + switch (*c) { + case '-': + add = 0; + c++; + break; + case '+': + c++; + default: + add = 1; + break; + } + addr = in_aton(c); + + spin_lock_bh(&recent_lock); + e = recent_entry_lookup(t, (const void *)&addr, NFPROTO_IPV4, 0); + if (e == NULL) { + if (add) + recent_entry_init(t, (const void *)&addr, + NFPROTO_IPV4, 0); + } else { + if (add) + recent_entry_update(t, e); + else + recent_entry_remove(t, e); + } + spin_unlock_bh(&recent_lock); + return size; +} + +static const struct file_operations recent_old_fops = { + .open = recent_old_seq_open, + .read = seq_read, + .write = recent_old_proc_write, + .release = seq_release_private, + .owner = THIS_MODULE, +}; +#endif + +static ssize_t +recent_mt_proc_write(struct file *file, const char __user *input, + size_t size, loff_t *loff) +{ + const struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode); + struct recent_table *t = pde->data; + struct recent_entry *e; + char buf[sizeof("+b335:1d35:1e55:dead:c0de:1715:5afe:c0de")]; + const char *c = buf; + union nf_inet_addr addr; + u_int16_t family; + bool add, succ; + + if (size == 0) + return 0; + if (size > sizeof(buf)) + size = sizeof(buf); + if (copy_from_user(buf, input, size) != 0) + return -EFAULT; + + /* Strict protocol! */ + if (*loff != 0) + return -ESPIPE; + switch (*c) { + case '/': /* flush table */ + spin_lock_bh(&recent_lock); + recent_table_flush(t); + spin_unlock_bh(&recent_lock); + return size; + case '-': /* remove address */ + add = false; + break; + case '+': /* add address */ + add = true; + break; + default: + printk(KERN_INFO KBUILD_MODNAME ": Need +ip, -ip or /\n"); + return -EINVAL; + } + + ++c; + --size; + if (strnchr(c, size, ':') != NULL) { + family = NFPROTO_IPV6; + succ = in6_pton(c, size, (void *)&addr, '\n', NULL); + } else { + family = NFPROTO_IPV4; + succ = in4_pton(c, size, (void *)&addr, '\n', NULL); + } + + if (!succ) { + printk(KERN_INFO KBUILD_MODNAME ": illegal address written " + "to procfs\n"); + return -EINVAL; + } + + spin_lock_bh(&recent_lock); + e = recent_entry_lookup(t, &addr, family, 0); + if (e == NULL) { + if (add) + recent_entry_init(t, &addr, family, 0); + } else { + if (add) + recent_entry_update(t, e); + else + recent_entry_remove(t, e); + } + spin_unlock_bh(&recent_lock); + /* Note we removed one above */ + *loff += size + 1; + return size + 1; +} + +static const struct file_operations recent_mt_fops = { + .open = recent_seq_open, + .read = seq_read, + .write = recent_mt_proc_write, + .release = seq_release_private, + .owner = THIS_MODULE, +}; +#endif /* CONFIG_PROC_FS */ + +static struct xt_match recent_mt_reg[] __read_mostly = { + { + .name = "recent", + .revision = 0, + .family = NFPROTO_IPV4, + .match = recent_mt, + .matchsize = sizeof(struct xt_recent_mtinfo), + .checkentry = recent_mt_check, + .destroy = recent_mt_destroy, + .me = THIS_MODULE, + }, + { + .name = "recent", + .revision = 0, + .family = NFPROTO_IPV6, + .match = recent_mt, + .matchsize = sizeof(struct xt_recent_mtinfo), + .checkentry = recent_mt_check, + .destroy = recent_mt_destroy, + .me = THIS_MODULE, + }, +}; + +static int __init recent_mt_init(void) +{ + int err; + + if (!ip_list_tot || !ip_pkt_list_tot || ip_pkt_list_tot > 255) + return -EINVAL; + ip_list_hash_size = 1 << fls(ip_list_tot); + + err = xt_register_matches(recent_mt_reg, ARRAY_SIZE(recent_mt_reg)); +#ifdef CONFIG_PROC_FS + if (err) + return err; + recent_proc_dir = proc_mkdir("xt_recent", init_net.proc_net); + if (recent_proc_dir == NULL) { + xt_unregister_matches(recent_mt_reg, ARRAY_SIZE(recent_mt_reg)); + err = -ENOMEM; + } +#ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT + if (err < 0) + return err; + proc_old_dir = proc_mkdir("ipt_recent", init_net.proc_net); + if (proc_old_dir == NULL) { + remove_proc_entry("xt_recent", init_net.proc_net); + xt_unregister_matches(recent_mt_reg, ARRAY_SIZE(recent_mt_reg)); + err = -ENOMEM; + } +#endif +#endif + return err; +} + +static void __exit recent_mt_exit(void) +{ + BUG_ON(!list_empty(&tables)); + xt_unregister_matches(recent_mt_reg, ARRAY_SIZE(recent_mt_reg)); +#ifdef CONFIG_PROC_FS +#ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT + remove_proc_entry("ipt_recent", init_net.proc_net); +#endif + remove_proc_entry("xt_recent", init_net.proc_net); +#endif +} + +module_init(recent_mt_init); +module_exit(recent_mt_exit); diff --git a/net/netfilter/xt_sctp.c b/net/netfilter/xt_sctp.c index e6e4681fa04..e223cb43ae8 100644 --- a/net/netfilter/xt_sctp.c +++ b/net/netfilter/xt_sctp.c @@ -117,23 +117,21 @@ match_packet(const struct sk_buff *skb, } static bool -sctp_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +sctp_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_sctp_info *info = matchinfo; + const struct xt_sctp_info *info = par->matchinfo; const sctp_sctphdr_t *sh; sctp_sctphdr_t _sh; - if (offset) { + if (par->fragoff != 0) { duprintf("Dropping non-first fragment.. FIXME\n"); return false; } - sh = skb_header_pointer(skb, protoff, sizeof(_sh), &_sh); + sh = skb_header_pointer(skb, par->thoff, sizeof(_sh), &_sh); if (sh == NULL) { duprintf("Dropping evil TCP offset=0 tinygram.\n"); - *hotdrop = true; + *par->hotdrop = true; return false; } duprintf("spt: %d\tdpt: %d\n", ntohs(sh->source), ntohs(sh->dest)); @@ -144,17 +142,14 @@ sctp_mt(const struct sk_buff *skb, const struct net_device *in, && SCCHECK(ntohs(sh->dest) >= info->dpts[0] && ntohs(sh->dest) <= info->dpts[1], XT_SCTP_DEST_PORTS, info->flags, info->invflags) - && SCCHECK(match_packet(skb, protoff + sizeof (sctp_sctphdr_t), - info, hotdrop), + && SCCHECK(match_packet(skb, par->thoff + sizeof(sctp_sctphdr_t), + info, par->hotdrop), XT_SCTP_CHUNK_TYPES, info->flags, info->invflags); } -static bool -sctp_mt_check(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool sctp_mt_check(const struct xt_mtchk_param *par) { - const struct xt_sctp_info *info = matchinfo; + const struct xt_sctp_info *info = par->matchinfo; return !(info->flags & ~XT_SCTP_VALID_FLAGS) && !(info->invflags & ~XT_SCTP_VALID_FLAGS) @@ -169,7 +164,7 @@ sctp_mt_check(const char *tablename, const void *inf, static struct xt_match sctp_mt_reg[] __read_mostly = { { .name = "sctp", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = sctp_mt_check, .match = sctp_mt, .matchsize = sizeof(struct xt_sctp_info), @@ -178,7 +173,7 @@ static struct xt_match sctp_mt_reg[] __read_mostly = { }, { .name = "sctp", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = sctp_mt_check, .match = sctp_mt, .matchsize = sizeof(struct xt_sctp_info), diff --git a/net/netfilter/xt_socket.c b/net/netfilter/xt_socket.c new file mode 100644 index 00000000000..02a8fed2108 --- /dev/null +++ b/net/netfilter/xt_socket.c @@ -0,0 +1,185 @@ +/* + * Transparent proxy support for Linux/iptables + * + * Copyright (C) 2007-2008 BalaBit IT Ltd. + * Author: Krisztian Kovacs + * + * 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/x_tables.h> +#include <linux/netfilter_ipv4/ip_tables.h> +#include <net/tcp.h> +#include <net/udp.h> +#include <net/icmp.h> +#include <net/sock.h> +#include <net/inet_sock.h> +#include <net/netfilter/nf_tproxy_core.h> +#include <net/netfilter/ipv4/nf_defrag_ipv4.h> + +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) +#define XT_SOCKET_HAVE_CONNTRACK 1 +#include <net/netfilter/nf_conntrack.h> +#endif + +static int +extract_icmp_fields(const struct sk_buff *skb, + u8 *protocol, + __be32 *raddr, + __be32 *laddr, + __be16 *rport, + __be16 *lport) +{ + unsigned int outside_hdrlen = ip_hdrlen(skb); + struct iphdr *inside_iph, _inside_iph; + struct icmphdr *icmph, _icmph; + __be16 *ports, _ports[2]; + + icmph = skb_header_pointer(skb, outside_hdrlen, + sizeof(_icmph), &_icmph); + if (icmph == NULL) + return 1; + + switch (icmph->type) { + case ICMP_DEST_UNREACH: + case ICMP_SOURCE_QUENCH: + case ICMP_REDIRECT: + case ICMP_TIME_EXCEEDED: + case ICMP_PARAMETERPROB: + break; + default: + return 1; + } + + inside_iph = skb_header_pointer(skb, outside_hdrlen + + sizeof(struct icmphdr), + sizeof(_inside_iph), &_inside_iph); + if (inside_iph == NULL) + return 1; + + if (inside_iph->protocol != IPPROTO_TCP && + inside_iph->protocol != IPPROTO_UDP) + return 1; + + ports = skb_header_pointer(skb, outside_hdrlen + + sizeof(struct icmphdr) + + (inside_iph->ihl << 2), + sizeof(_ports), &_ports); + if (ports == NULL) + return 1; + + /* the inside IP packet is the one quoted from our side, thus + * its saddr is the local address */ + *protocol = inside_iph->protocol; + *laddr = inside_iph->saddr; + *lport = ports[0]; + *raddr = inside_iph->daddr; + *rport = ports[1]; + + return 0; +} + + +static bool +socket_mt(const struct sk_buff *skb, const struct xt_match_param *par) +{ + const struct iphdr *iph = ip_hdr(skb); + struct udphdr _hdr, *hp = NULL; + struct sock *sk; + __be32 daddr, saddr; + __be16 dport, sport; + u8 protocol; +#ifdef XT_SOCKET_HAVE_CONNTRACK + struct nf_conn const *ct; + enum ip_conntrack_info ctinfo; +#endif + + if (iph->protocol == IPPROTO_UDP || iph->protocol == IPPROTO_TCP) { + hp = skb_header_pointer(skb, ip_hdrlen(skb), + sizeof(_hdr), &_hdr); + if (hp == NULL) + return false; + + protocol = iph->protocol; + saddr = iph->saddr; + sport = hp->source; + daddr = iph->daddr; + dport = hp->dest; + + } else if (iph->protocol == IPPROTO_ICMP) { + if (extract_icmp_fields(skb, &protocol, &saddr, &daddr, + &sport, &dport)) + return false; + } else { + return false; + } + +#ifdef XT_SOCKET_HAVE_CONNTRACK + /* Do the lookup with the original socket address in case this is a + * reply packet of an established SNAT-ted connection. */ + + ct = nf_ct_get(skb, &ctinfo); + if (ct && (ct != &nf_conntrack_untracked) && + ((iph->protocol != IPPROTO_ICMP && + ctinfo == IP_CT_IS_REPLY + IP_CT_ESTABLISHED) || + (iph->protocol == IPPROTO_ICMP && + ctinfo == IP_CT_IS_REPLY + IP_CT_RELATED)) && + (ct->status & IPS_SRC_NAT_DONE)) { + + daddr = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip; + dport = (iph->protocol == IPPROTO_TCP) ? + ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.tcp.port : + ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.udp.port; + } +#endif + + sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), protocol, + saddr, daddr, sport, dport, par->in, false); + if (sk != NULL) { + bool wildcard = (inet_sk(sk)->rcv_saddr == 0); + + nf_tproxy_put_sock(sk); + if (wildcard) + sk = NULL; + } + + pr_debug("socket match: proto %u %08x:%u -> %08x:%u " + "(orig %08x:%u) sock %p\n", + protocol, ntohl(saddr), ntohs(sport), + ntohl(daddr), ntohs(dport), + ntohl(iph->daddr), hp ? ntohs(hp->dest) : 0, sk); + + return (sk != NULL); +} + +static struct xt_match socket_mt_reg __read_mostly = { + .name = "socket", + .family = AF_INET, + .match = socket_mt, + .hooks = 1 << NF_INET_PRE_ROUTING, + .me = THIS_MODULE, +}; + +static int __init socket_mt_init(void) +{ + nf_defrag_ipv4_enable(); + return xt_register_match(&socket_mt_reg); +} + +static void __exit socket_mt_exit(void) +{ + xt_unregister_match(&socket_mt_reg); +} + +module_init(socket_mt_init); +module_exit(socket_mt_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Krisztian Kovacs, Balazs Scheidler"); +MODULE_DESCRIPTION("x_tables socket match module"); +MODULE_ALIAS("ipt_socket"); diff --git a/net/netfilter/xt_state.c b/net/netfilter/xt_state.c index a776dc36a19..4c946cbd731 100644 --- a/net/netfilter/xt_state.c +++ b/net/netfilter/xt_state.c @@ -21,12 +21,9 @@ MODULE_ALIAS("ipt_state"); MODULE_ALIAS("ip6t_state"); static bool -state_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +state_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_state_info *sinfo = matchinfo; + const struct xt_state_info *sinfo = par->matchinfo; enum ip_conntrack_info ctinfo; unsigned int statebit; @@ -40,28 +37,25 @@ state_mt(const struct sk_buff *skb, const struct net_device *in, return (sinfo->statemask & statebit); } -static bool -state_mt_check(const char *tablename, const void *inf, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool state_mt_check(const struct xt_mtchk_param *par) { - if (nf_ct_l3proto_try_module_get(match->family) < 0) { + if (nf_ct_l3proto_try_module_get(par->match->family) < 0) { printk(KERN_WARNING "can't load conntrack support for " - "proto=%u\n", match->family); + "proto=%u\n", par->match->family); return false; } return true; } -static void state_mt_destroy(const struct xt_match *match, void *matchinfo) +static void state_mt_destroy(const struct xt_mtdtor_param *par) { - nf_ct_l3proto_module_put(match->family); + nf_ct_l3proto_module_put(par->match->family); } static struct xt_match state_mt_reg[] __read_mostly = { { .name = "state", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = state_mt_check, .match = state_mt, .destroy = state_mt_destroy, @@ -70,7 +64,7 @@ static struct xt_match state_mt_reg[] __read_mostly = { }, { .name = "state", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = state_mt_check, .match = state_mt, .destroy = state_mt_destroy, diff --git a/net/netfilter/xt_statistic.c b/net/netfilter/xt_statistic.c index 43133080da7..0d75141139d 100644 --- a/net/netfilter/xt_statistic.c +++ b/net/netfilter/xt_statistic.c @@ -25,12 +25,9 @@ MODULE_ALIAS("ip6t_statistic"); static DEFINE_SPINLOCK(nth_lock); static bool -statistic_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +statistic_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - struct xt_statistic_info *info = (struct xt_statistic_info *)matchinfo; + struct xt_statistic_info *info = (void *)par->matchinfo; bool ret = info->flags & XT_STATISTIC_INVERT; switch (info->mode) { @@ -52,12 +49,9 @@ statistic_mt(const struct sk_buff *skb, const struct net_device *in, return ret; } -static bool -statistic_mt_check(const char *tablename, const void *entry, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool statistic_mt_check(const struct xt_mtchk_param *par) { - struct xt_statistic_info *info = matchinfo; + struct xt_statistic_info *info = par->matchinfo; if (info->mode > XT_STATISTIC_MODE_MAX || info->flags & ~XT_STATISTIC_MASK) @@ -66,35 +60,24 @@ statistic_mt_check(const char *tablename, const void *entry, return true; } -static struct xt_match statistic_mt_reg[] __read_mostly = { - { - .name = "statistic", - .family = AF_INET, - .checkentry = statistic_mt_check, - .match = statistic_mt, - .matchsize = sizeof(struct xt_statistic_info), - .me = THIS_MODULE, - }, - { - .name = "statistic", - .family = AF_INET6, - .checkentry = statistic_mt_check, - .match = statistic_mt, - .matchsize = sizeof(struct xt_statistic_info), - .me = THIS_MODULE, - }, +static struct xt_match xt_statistic_mt_reg __read_mostly = { + .name = "statistic", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = statistic_mt, + .checkentry = statistic_mt_check, + .matchsize = sizeof(struct xt_statistic_info), + .me = THIS_MODULE, }; static int __init statistic_mt_init(void) { - return xt_register_matches(statistic_mt_reg, - ARRAY_SIZE(statistic_mt_reg)); + return xt_register_match(&xt_statistic_mt_reg); } static void __exit statistic_mt_exit(void) { - xt_unregister_matches(statistic_mt_reg, - ARRAY_SIZE(statistic_mt_reg)); + xt_unregister_match(&xt_statistic_mt_reg); } module_init(statistic_mt_init); diff --git a/net/netfilter/xt_string.c b/net/netfilter/xt_string.c index 4903182a062..b4d77411131 100644 --- a/net/netfilter/xt_string.c +++ b/net/netfilter/xt_string.c @@ -22,18 +22,15 @@ MODULE_ALIAS("ipt_string"); MODULE_ALIAS("ip6t_string"); static bool -string_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +string_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_string_info *conf = matchinfo; + const struct xt_string_info *conf = par->matchinfo; struct ts_state state; int invert; memset(&state, 0, sizeof(struct ts_state)); - invert = (match->revision == 0 ? conf->u.v0.invert : + invert = (par->match->revision == 0 ? conf->u.v0.invert : conf->u.v1.flags & XT_STRING_FLAG_INVERT); return (skb_find_text((struct sk_buff *)skb, conf->from_offset, @@ -43,12 +40,9 @@ string_mt(const struct sk_buff *skb, const struct net_device *in, #define STRING_TEXT_PRIV(m) ((struct xt_string_info *)(m)) -static bool -string_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool string_mt_check(const struct xt_mtchk_param *par) { - struct xt_string_info *conf = matchinfo; + struct xt_string_info *conf = par->matchinfo; struct ts_config *ts_conf; int flags = TS_AUTOLOAD; @@ -59,7 +53,7 @@ string_mt_check(const char *tablename, const void *ip, return false; if (conf->patlen > XT_STRING_MAX_PATTERN_SIZE) return false; - if (match->revision == 1) { + if (par->match->revision == 1) { if (conf->u.v1.flags & ~(XT_STRING_FLAG_IGNORECASE | XT_STRING_FLAG_INVERT)) return false; @@ -76,36 +70,16 @@ string_mt_check(const char *tablename, const void *ip, return true; } -static void string_mt_destroy(const struct xt_match *match, void *matchinfo) +static void string_mt_destroy(const struct xt_mtdtor_param *par) { - textsearch_destroy(STRING_TEXT_PRIV(matchinfo)->config); + textsearch_destroy(STRING_TEXT_PRIV(par->matchinfo)->config); } -static struct xt_match string_mt_reg[] __read_mostly = { - { - .name = "string", - .revision = 0, - .family = AF_INET, - .checkentry = string_mt_check, - .match = string_mt, - .destroy = string_mt_destroy, - .matchsize = sizeof(struct xt_string_info), - .me = THIS_MODULE - }, - { - .name = "string", - .revision = 1, - .family = AF_INET, - .checkentry = string_mt_check, - .match = string_mt, - .destroy = string_mt_destroy, - .matchsize = sizeof(struct xt_string_info), - .me = THIS_MODULE - }, +static struct xt_match xt_string_mt_reg[] __read_mostly = { { .name = "string", .revision = 0, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .checkentry = string_mt_check, .match = string_mt, .destroy = string_mt_destroy, @@ -115,7 +89,7 @@ static struct xt_match string_mt_reg[] __read_mostly = { { .name = "string", .revision = 1, - .family = AF_INET6, + .family = NFPROTO_UNSPEC, .checkentry = string_mt_check, .match = string_mt, .destroy = string_mt_destroy, @@ -126,12 +100,13 @@ static struct xt_match string_mt_reg[] __read_mostly = { static int __init string_mt_init(void) { - return xt_register_matches(string_mt_reg, ARRAY_SIZE(string_mt_reg)); + return xt_register_matches(xt_string_mt_reg, + ARRAY_SIZE(xt_string_mt_reg)); } static void __exit string_mt_exit(void) { - xt_unregister_matches(string_mt_reg, ARRAY_SIZE(string_mt_reg)); + xt_unregister_matches(xt_string_mt_reg, ARRAY_SIZE(xt_string_mt_reg)); } module_init(string_mt_init); diff --git a/net/netfilter/xt_tcpmss.c b/net/netfilter/xt_tcpmss.c index 6771bf01275..4809b34b10f 100644 --- a/net/netfilter/xt_tcpmss.c +++ b/net/netfilter/xt_tcpmss.c @@ -25,12 +25,9 @@ MODULE_ALIAS("ipt_tcpmss"); MODULE_ALIAS("ip6t_tcpmss"); static bool -tcpmss_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, - bool *hotdrop) +tcpmss_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_tcpmss_match_info *info = matchinfo; + const struct xt_tcpmss_match_info *info = par->matchinfo; const struct tcphdr *th; struct tcphdr _tcph; /* tcp.doff is only 4 bits, ie. max 15 * 4 bytes */ @@ -39,7 +36,7 @@ tcpmss_mt(const struct sk_buff *skb, const struct net_device *in, unsigned int i, optlen; /* If we don't have the whole header, drop packet. */ - th = skb_header_pointer(skb, protoff, sizeof(_tcph), &_tcph); + th = skb_header_pointer(skb, par->thoff, sizeof(_tcph), &_tcph); if (th == NULL) goto dropit; @@ -52,7 +49,7 @@ tcpmss_mt(const struct sk_buff *skb, const struct net_device *in, goto out; /* Truncated options. */ - op = skb_header_pointer(skb, protoff + sizeof(*th), optlen, _opt); + op = skb_header_pointer(skb, par->thoff + sizeof(*th), optlen, _opt); if (op == NULL) goto dropit; @@ -76,14 +73,14 @@ out: return info->invert; dropit: - *hotdrop = true; + *par->hotdrop = true; return false; } static struct xt_match tcpmss_mt_reg[] __read_mostly = { { .name = "tcpmss", - .family = AF_INET, + .family = NFPROTO_IPV4, .match = tcpmss_mt, .matchsize = sizeof(struct xt_tcpmss_match_info), .proto = IPPROTO_TCP, @@ -91,7 +88,7 @@ static struct xt_match tcpmss_mt_reg[] __read_mostly = { }, { .name = "tcpmss", - .family = AF_INET6, + .family = NFPROTO_IPV6, .match = tcpmss_mt, .matchsize = sizeof(struct xt_tcpmss_match_info), .proto = IPPROTO_TCP, diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c index 951b06b8d70..1ebdc4934ee 100644 --- a/net/netfilter/xt_tcpudp.c +++ b/net/netfilter/xt_tcpudp.c @@ -68,25 +68,22 @@ tcp_find_option(u_int8_t option, return invert; } -static bool -tcp_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +static bool tcp_mt(const struct sk_buff *skb, const struct xt_match_param *par) { const struct tcphdr *th; struct tcphdr _tcph; - const struct xt_tcp *tcpinfo = matchinfo; + const struct xt_tcp *tcpinfo = par->matchinfo; - if (offset) { + if (par->fragoff != 0) { /* 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) { + if (par->fragoff == 1) { duprintf("Dropping evil TCP offset=1 frag.\n"); - *hotdrop = true; + *par->hotdrop = true; } /* Must not be a fragment. */ return false; @@ -94,12 +91,12 @@ tcp_mt(const struct sk_buff *skb, const struct net_device *in, #define FWINVTCP(bool, invflg) ((bool) ^ !!(tcpinfo->invflags & (invflg))) - th = skb_header_pointer(skb, protoff, sizeof(_tcph), &_tcph); + th = skb_header_pointer(skb, par->thoff, 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 = true; + *par->hotdrop = true; return false; } @@ -117,49 +114,42 @@ tcp_mt(const struct sk_buff *skb, const struct net_device *in, return false; if (tcpinfo->option) { if (th->doff * 4 < sizeof(_tcph)) { - *hotdrop = true; + *par->hotdrop = true; return false; } - if (!tcp_find_option(tcpinfo->option, skb, protoff, + if (!tcp_find_option(tcpinfo->option, skb, par->thoff, th->doff*4 - sizeof(_tcph), tcpinfo->invflags & XT_TCP_INV_OPTION, - hotdrop)) + par->hotdrop)) return false; } return true; } -/* Called when user tries to insert an entry of this type. */ -static bool -tcp_mt_check(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool tcp_mt_check(const struct xt_mtchk_param *par) { - const struct xt_tcp *tcpinfo = matchinfo; + const struct xt_tcp *tcpinfo = par->matchinfo; /* Must specify no unknown invflags */ return !(tcpinfo->invflags & ~XT_TCP_INV_MASK); } -static bool -udp_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +static bool udp_mt(const struct sk_buff *skb, const struct xt_match_param *par) { const struct udphdr *uh; struct udphdr _udph; - const struct xt_udp *udpinfo = matchinfo; + const struct xt_udp *udpinfo = par->matchinfo; /* Must not be a fragment. */ - if (offset) + if (par->fragoff != 0) return false; - uh = skb_header_pointer(skb, protoff, sizeof(_udph), &_udph); + uh = skb_header_pointer(skb, par->thoff, 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 = true; + *par->hotdrop = true; return false; } @@ -171,13 +161,9 @@ udp_mt(const struct sk_buff *skb, const struct net_device *in, !!(udpinfo->invflags & XT_UDP_INV_DSTPT)); } -/* Called when user tries to insert an entry of this type. */ -static bool -udp_mt_check(const char *tablename, const void *info, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool udp_mt_check(const struct xt_mtchk_param *par) { - const struct xt_udp *udpinfo = matchinfo; + const struct xt_udp *udpinfo = par->matchinfo; /* Must specify no unknown invflags */ return !(udpinfo->invflags & ~XT_UDP_INV_MASK); @@ -186,7 +172,7 @@ udp_mt_check(const char *tablename, const void *info, static struct xt_match tcpudp_mt_reg[] __read_mostly = { { .name = "tcp", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = tcp_mt_check, .match = tcp_mt, .matchsize = sizeof(struct xt_tcp), @@ -195,7 +181,7 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = { }, { .name = "tcp", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = tcp_mt_check, .match = tcp_mt, .matchsize = sizeof(struct xt_tcp), @@ -204,7 +190,7 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = { }, { .name = "udp", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = udp_mt_check, .match = udp_mt, .matchsize = sizeof(struct xt_udp), @@ -213,7 +199,7 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = { }, { .name = "udp", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = udp_mt_check, .match = udp_mt, .matchsize = sizeof(struct xt_udp), @@ -222,7 +208,7 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = { }, { .name = "udplite", - .family = AF_INET, + .family = NFPROTO_IPV4, .checkentry = udp_mt_check, .match = udp_mt, .matchsize = sizeof(struct xt_udp), @@ -231,7 +217,7 @@ static struct xt_match tcpudp_mt_reg[] __read_mostly = { }, { .name = "udplite", - .family = AF_INET6, + .family = NFPROTO_IPV6, .checkentry = udp_mt_check, .match = udp_mt, .matchsize = sizeof(struct xt_udp), diff --git a/net/netfilter/xt_time.c b/net/netfilter/xt_time.c index 9f328593287..29375ba8db7 100644 --- a/net/netfilter/xt_time.c +++ b/net/netfilter/xt_time.c @@ -136,26 +136,26 @@ static void localtime_3(struct xtm *r, time_t time) * from w repeatedly while counting.) */ if (is_leap(year)) { + /* use days_since_leapyear[] in a leap year */ for (i = ARRAY_SIZE(days_since_leapyear) - 1; - i > 0 && days_since_year[i] > w; --i) + i > 0 && days_since_leapyear[i] > w; --i) /* just loop */; + r->monthday = w - days_since_leapyear[i] + 1; } else { for (i = ARRAY_SIZE(days_since_year) - 1; i > 0 && days_since_year[i] > w; --i) /* just loop */; + r->monthday = w - days_since_year[i] + 1; } r->month = i + 1; - r->monthday = w - days_since_year[i] + 1; return; } static bool -time_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +time_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_time_info *info = matchinfo; + const struct xt_time_info *info = par->matchinfo; unsigned int packet_time; struct xtm current_time; s64 stamp; @@ -218,12 +218,9 @@ time_mt(const struct sk_buff *skb, const struct net_device *in, return true; } -static bool -time_mt_check(const char *tablename, const void *ip, - const struct xt_match *match, void *matchinfo, - unsigned int hook_mask) +static bool time_mt_check(const struct xt_mtchk_param *par) { - const struct xt_time_info *info = matchinfo; + const struct xt_time_info *info = par->matchinfo; if (info->daytime_start > XT_TIME_MAX_DAYTIME || info->daytime_stop > XT_TIME_MAX_DAYTIME) { @@ -235,33 +232,23 @@ time_mt_check(const char *tablename, const void *ip, return true; } -static struct xt_match time_mt_reg[] __read_mostly = { - { - .name = "time", - .family = AF_INET, - .match = time_mt, - .matchsize = sizeof(struct xt_time_info), - .checkentry = time_mt_check, - .me = THIS_MODULE, - }, - { - .name = "time", - .family = AF_INET6, - .match = time_mt, - .matchsize = sizeof(struct xt_time_info), - .checkentry = time_mt_check, - .me = THIS_MODULE, - }, +static struct xt_match xt_time_mt_reg __read_mostly = { + .name = "time", + .family = NFPROTO_UNSPEC, + .match = time_mt, + .checkentry = time_mt_check, + .matchsize = sizeof(struct xt_time_info), + .me = THIS_MODULE, }; static int __init time_mt_init(void) { - return xt_register_matches(time_mt_reg, ARRAY_SIZE(time_mt_reg)); + return xt_register_match(&xt_time_mt_reg); } static void __exit time_mt_exit(void) { - xt_unregister_matches(time_mt_reg, ARRAY_SIZE(time_mt_reg)); + xt_unregister_match(&xt_time_mt_reg); } module_init(time_mt_init); diff --git a/net/netfilter/xt_u32.c b/net/netfilter/xt_u32.c index 627e0f336d5..24a52762450 100644 --- a/net/netfilter/xt_u32.c +++ b/net/netfilter/xt_u32.c @@ -87,43 +87,32 @@ static bool u32_match_it(const struct xt_u32 *data, return true; } -static bool -u32_mt(const struct sk_buff *skb, const struct net_device *in, - const struct net_device *out, const struct xt_match *match, - const void *matchinfo, int offset, unsigned int protoff, bool *hotdrop) +static bool u32_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - const struct xt_u32 *data = matchinfo; + const struct xt_u32 *data = par->matchinfo; bool ret; ret = u32_match_it(data, skb); return ret ^ data->invert; } -static struct xt_match u32_mt_reg[] __read_mostly = { - { - .name = "u32", - .family = AF_INET, - .match = u32_mt, - .matchsize = sizeof(struct xt_u32), - .me = THIS_MODULE, - }, - { - .name = "u32", - .family = AF_INET6, - .match = u32_mt, - .matchsize = sizeof(struct xt_u32), - .me = THIS_MODULE, - }, +static struct xt_match xt_u32_mt_reg __read_mostly = { + .name = "u32", + .revision = 0, + .family = NFPROTO_UNSPEC, + .match = u32_mt, + .matchsize = sizeof(struct xt_u32), + .me = THIS_MODULE, }; static int __init u32_mt_init(void) { - return xt_register_matches(u32_mt_reg, ARRAY_SIZE(u32_mt_reg)); + return xt_register_match(&xt_u32_mt_reg); } static void __exit u32_mt_exit(void) { - xt_unregister_matches(u32_mt_reg, ARRAY_SIZE(u32_mt_reg)); + xt_unregister_match(&xt_u32_mt_reg); } module_init(u32_mt_init); |