diff options
Diffstat (limited to 'net/sched')
79 files changed, 51575 insertions, 0 deletions
diff --git a/net/sched/Kconfig b/net/sched/Kconfig new file mode 100644 index 000000000..e95741388 --- /dev/null +++ b/net/sched/Kconfig @@ -0,0 +1,927 @@ +# +# Traffic control configuration. +# + +menuconfig NET_SCHED + bool "QoS and/or fair queueing" + select NET_SCH_FIFO + ---help--- + When the kernel has several packets to send out over a network + device, it has to decide which ones to send first, which ones to + delay, and which ones to drop. This is the job of the queueing + disciplines, several different algorithms for how to do this + "fairly" have been proposed. + + If you say N here, you will get the standard packet scheduler, which + is a FIFO (first come, first served). If you say Y here, you will be + able to choose from among several alternative algorithms which can + then be attached to different network devices. This is useful for + example if some of your network devices are real time devices that + need a certain minimum data flow rate, or if you need to limit the + maximum data flow rate for traffic which matches specified criteria. + This code is considered to be experimental. + + To administer these schedulers, you'll need the user-level utilities + from the package iproute2+tc at + <https://www.kernel.org/pub/linux/utils/net/iproute2/>. That package + also contains some documentation; for more, check out + <http://www.linuxfoundation.org/collaborate/workgroups/networking/iproute2>. + + This Quality of Service (QoS) support will enable you to use + Differentiated Services (diffserv) and Resource Reservation Protocol + (RSVP) on your Linux router if you also say Y to the corresponding + classifiers below. Documentation and software is at + <http://diffserv.sourceforge.net/>. + + If you say Y here and to "/proc file system" below, you will be able + to read status information about packet schedulers from the file + /proc/net/psched. + + The available schedulers are listed in the following questions; you + can say Y to as many as you like. If unsure, say N now. + +if NET_SCHED + +comment "Queueing/Scheduling" + +config NET_SCH_CBQ + tristate "Class Based Queueing (CBQ)" + ---help--- + Say Y here if you want to use the Class-Based Queueing (CBQ) packet + scheduling algorithm. This algorithm classifies the waiting packets + into a tree-like hierarchy of classes; the leaves of this tree are + in turn scheduled by separate algorithms. + + See the top of <file:net/sched/sch_cbq.c> for more details. + + CBQ is a commonly used scheduler, so if you're unsure, you should + say Y here. Then say Y to all the queueing algorithms below that you + want to use as leaf disciplines. + + To compile this code as a module, choose M here: the + module will be called sch_cbq. + +config NET_SCH_HTB + tristate "Hierarchical Token Bucket (HTB)" + ---help--- + Say Y here if you want to use the Hierarchical Token Buckets (HTB) + packet scheduling algorithm. See + <http://luxik.cdi.cz/~devik/qos/htb/> for complete manual and + in-depth articles. + + HTB is very similar to CBQ regarding its goals however is has + different properties and different algorithm. + + To compile this code as a module, choose M here: the + module will be called sch_htb. + +config NET_SCH_HFSC + tristate "Hierarchical Fair Service Curve (HFSC)" + ---help--- + Say Y here if you want to use the Hierarchical Fair Service Curve + (HFSC) packet scheduling algorithm. + + To compile this code as a module, choose M here: the + module will be called sch_hfsc. + +config NET_SCH_ATM + tristate "ATM Virtual Circuits (ATM)" + depends on ATM + ---help--- + Say Y here if you want to use the ATM pseudo-scheduler. This + provides a framework for invoking classifiers, which in turn + select classes of this queuing discipline. Each class maps + the flow(s) it is handling to a given virtual circuit. + + See the top of <file:net/sched/sch_atm.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_atm. + +config NET_SCH_PRIO + tristate "Multi Band Priority Queueing (PRIO)" + ---help--- + Say Y here if you want to use an n-band priority queue packet + scheduler. + + To compile this code as a module, choose M here: the + module will be called sch_prio. + +config NET_SCH_MULTIQ + tristate "Hardware Multiqueue-aware Multi Band Queuing (MULTIQ)" + ---help--- + Say Y here if you want to use an n-band queue packet scheduler + to support devices that have multiple hardware transmit queues. + + To compile this code as a module, choose M here: the + module will be called sch_multiq. + +config NET_SCH_RED + tristate "Random Early Detection (RED)" + ---help--- + Say Y here if you want to use the Random Early Detection (RED) + packet scheduling algorithm. + + See the top of <file:net/sched/sch_red.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_red. + +config NET_SCH_SFB + tristate "Stochastic Fair Blue (SFB)" + ---help--- + Say Y here if you want to use the Stochastic Fair Blue (SFB) + packet scheduling algorithm. + + See the top of <file:net/sched/sch_sfb.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_sfb. + +config NET_SCH_SFQ + tristate "Stochastic Fairness Queueing (SFQ)" + ---help--- + Say Y here if you want to use the Stochastic Fairness Queueing (SFQ) + packet scheduling algorithm. + + See the top of <file:net/sched/sch_sfq.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_sfq. + +config NET_SCH_TEQL + tristate "True Link Equalizer (TEQL)" + ---help--- + Say Y here if you want to use the True Link Equalizer (TLE) packet + scheduling algorithm. This queueing discipline allows the combination + of several physical devices into one virtual device. + + See the top of <file:net/sched/sch_teql.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_teql. + +config NET_SCH_TBF + tristate "Token Bucket Filter (TBF)" + ---help--- + Say Y here if you want to use the Token Bucket Filter (TBF) packet + scheduling algorithm. + + See the top of <file:net/sched/sch_tbf.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_tbf. + +config NET_SCH_CBS + tristate "Credit Based Shaper (CBS)" + ---help--- + Say Y here if you want to use the Credit Based Shaper (CBS) packet + scheduling algorithm. + + See the top of <file:net/sched/sch_cbs.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_cbs. + +config NET_SCH_ETF + tristate "Earliest TxTime First (ETF)" + help + Say Y here if you want to use the Earliest TxTime First (ETF) packet + scheduling algorithm. + + See the top of <file:net/sched/sch_etf.c> for more details. + + To compile this code as a module, choose M here: the + module will be called sch_etf. + +config NET_SCH_GRED + tristate "Generic Random Early Detection (GRED)" + ---help--- + Say Y here if you want to use the Generic Random Early Detection + (GRED) packet scheduling algorithm for some of your network devices + (see the top of <file:net/sched/sch_red.c> for details and + references about the algorithm). + + To compile this code as a module, choose M here: the + module will be called sch_gred. + +config NET_SCH_DSMARK + tristate "Differentiated Services marker (DSMARK)" + ---help--- + Say Y if you want to schedule packets according to the + Differentiated Services architecture proposed in RFC 2475. + Technical information on this method, with pointers to associated + RFCs, is available at <http://www.gta.ufrj.br/diffserv/>. + + To compile this code as a module, choose M here: the + module will be called sch_dsmark. + +config NET_SCH_NETEM + tristate "Network emulator (NETEM)" + ---help--- + Say Y if you want to emulate network delay, loss, and packet + re-ordering. This is often useful to simulate networks when + testing applications or protocols. + + To compile this driver as a module, choose M here: the module + will be called sch_netem. + + If unsure, say N. + +config NET_SCH_DRR + tristate "Deficit Round Robin scheduler (DRR)" + help + Say Y here if you want to use the Deficit Round Robin (DRR) packet + scheduling algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_drr. + + If unsure, say N. + +config NET_SCH_MQPRIO + tristate "Multi-queue priority scheduler (MQPRIO)" + help + Say Y here if you want to use the Multi-queue Priority scheduler. + This scheduler allows QOS to be offloaded on NICs that have support + for offloading QOS schedulers. + + To compile this driver as a module, choose M here: the module will + be called sch_mqprio. + + If unsure, say N. + +config NET_SCH_SKBPRIO + tristate "SKB priority queue scheduler (SKBPRIO)" + help + Say Y here if you want to use the SKB priority queue + scheduler. This schedules packets according to skb->priority, + which is useful for request packets in DoS mitigation systems such + as Gatekeeper. + + To compile this driver as a module, choose M here: the module will + be called sch_skbprio. + + If unsure, say N. + +config NET_SCH_CHOKE + tristate "CHOose and Keep responsive flow scheduler (CHOKE)" + help + Say Y here if you want to use the CHOKe packet scheduler (CHOose + and Keep for responsive flows, CHOose and Kill for unresponsive + flows). This is a variation of RED which trys to penalize flows + that monopolize the queue. + + To compile this code as a module, choose M here: the + module will be called sch_choke. + +config NET_SCH_QFQ + tristate "Quick Fair Queueing scheduler (QFQ)" + help + Say Y here if you want to use the Quick Fair Queueing Scheduler (QFQ) + packet scheduling algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_qfq. + + If unsure, say N. + +config NET_SCH_CODEL + tristate "Controlled Delay AQM (CODEL)" + help + Say Y here if you want to use the Controlled Delay (CODEL) + packet scheduling algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_codel. + + If unsure, say N. + +config NET_SCH_FQ_CODEL + tristate "Fair Queue Controlled Delay AQM (FQ_CODEL)" + help + Say Y here if you want to use the FQ Controlled Delay (FQ_CODEL) + packet scheduling algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_fq_codel. + + If unsure, say N. + +config NET_SCH_CAKE + tristate "Common Applications Kept Enhanced (CAKE)" + help + Say Y here if you want to use the Common Applications Kept Enhanced + (CAKE) queue management algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_cake. + + If unsure, say N. + +config NET_SCH_FQ + tristate "Fair Queue" + help + Say Y here if you want to use the FQ packet scheduling algorithm. + + FQ does flow separation, and is able to respect pacing requirements + set by TCP stack into sk->sk_pacing_rate (for localy generated + traffic) + + To compile this driver as a module, choose M here: the module + will be called sch_fq. + + If unsure, say N. + +config NET_SCH_HHF + tristate "Heavy-Hitter Filter (HHF)" + help + Say Y here if you want to use the Heavy-Hitter Filter (HHF) + packet scheduling algorithm. + + To compile this driver as a module, choose M here: the module + will be called sch_hhf. + +config NET_SCH_PIE + tristate "Proportional Integral controller Enhanced (PIE) scheduler" + help + Say Y here if you want to use the Proportional Integral controller + Enhanced scheduler packet scheduling algorithm. + For more information, please see + http://tools.ietf.org/html/draft-pan-tsvwg-pie-00 + + To compile this driver as a module, choose M here: the module + will be called sch_pie. + + If unsure, say N. + +config NET_SCH_INGRESS + tristate "Ingress/classifier-action Qdisc" + depends on NET_CLS_ACT + select NET_INGRESS + select NET_EGRESS + ---help--- + Say Y here if you want to use classifiers for incoming and/or outgoing + packets. This qdisc doesn't do anything else besides running classifiers, + which can also have actions attached to them. In case of outgoing packets, + classifiers that this qdisc holds are executed in the transmit path + before real enqueuing to an egress qdisc happens. + + If unsure, say Y. + + To compile this code as a module, choose M here: the module will be + called sch_ingress with alias of sch_clsact. + +config NET_SCH_PLUG + tristate "Plug network traffic until release (PLUG)" + ---help--- + + This queuing discipline allows userspace to plug/unplug a network + output queue, using the netlink interface. When it receives an + enqueue command it inserts a plug into the outbound queue that + causes following packets to enqueue until a dequeue command arrives + over netlink, causing the plug to be removed and resuming the normal + packet flow. + + This module also provides a generic "network output buffering" + functionality (aka output commit), wherein upon arrival of a dequeue + command, only packets up to the first plug are released for delivery. + The Remus HA project uses this module to enable speculative execution + of virtual machines by allowing the generated network output to be rolled + back if needed. + + For more information, please refer to <http://wiki.xenproject.org/wiki/Remus> + + Say Y here if you are using this kernel for Xen dom0 and + want to protect Xen guests with Remus. + + To compile this code as a module, choose M here: the + module will be called sch_plug. + +menuconfig NET_SCH_DEFAULT + bool "Allow override default queue discipline" + ---help--- + Support for selection of default queuing discipline. + + Nearly all users can safely say no here, and the default + of pfifo_fast will be used. Many distributions already set + the default value via /proc/sys/net/core/default_qdisc. + + If unsure, say N. + +if NET_SCH_DEFAULT + +choice + prompt "Default queuing discipline" + default DEFAULT_PFIFO_FAST + help + Select the queueing discipline that will be used by default + for all network devices. + + config DEFAULT_FQ + bool "Fair Queue" if NET_SCH_FQ + + config DEFAULT_CODEL + bool "Controlled Delay" if NET_SCH_CODEL + + config DEFAULT_FQ_CODEL + bool "Fair Queue Controlled Delay" if NET_SCH_FQ_CODEL + + config DEFAULT_SFQ + bool "Stochastic Fair Queue" if NET_SCH_SFQ + + config DEFAULT_PFIFO_FAST + bool "Priority FIFO Fast" +endchoice + +config DEFAULT_NET_SCH + string + default "pfifo_fast" if DEFAULT_PFIFO_FAST + default "fq" if DEFAULT_FQ + default "fq_codel" if DEFAULT_FQ_CODEL + default "sfq" if DEFAULT_SFQ + default "pfifo_fast" +endif + +comment "Classification" + +config NET_CLS + bool + +config NET_CLS_BASIC + tristate "Elementary classification (BASIC)" + select NET_CLS + ---help--- + Say Y here if you want to be able to classify packets using + only extended matches and actions. + + To compile this code as a module, choose M here: the + module will be called cls_basic. + +config NET_CLS_TCINDEX + tristate "Traffic-Control Index (TCINDEX)" + select NET_CLS + ---help--- + Say Y here if you want to be able to classify packets based on + traffic control indices. You will want this feature if you want + to implement Differentiated Services together with DSMARK. + + To compile this code as a module, choose M here: the + module will be called cls_tcindex. + +config NET_CLS_ROUTE4 + tristate "Routing decision (ROUTE)" + depends on INET + select IP_ROUTE_CLASSID + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets + according to the route table entry they matched. + + To compile this code as a module, choose M here: the + module will be called cls_route. + +config NET_CLS_FW + tristate "Netfilter mark (FW)" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets + according to netfilter/firewall marks. + + To compile this code as a module, choose M here: the + module will be called cls_fw. + +config NET_CLS_U32 + tristate "Universal 32bit comparisons w/ hashing (U32)" + select NET_CLS + ---help--- + Say Y here to be able to classify packets using a universal + 32bit pieces based comparison scheme. + + To compile this code as a module, choose M here: the + module will be called cls_u32. + +config CLS_U32_PERF + bool "Performance counters support" + depends on NET_CLS_U32 + ---help--- + Say Y here to make u32 gather additional statistics useful for + fine tuning u32 classifiers. + +config CLS_U32_MARK + bool "Netfilter marks support" + depends on NET_CLS_U32 + ---help--- + Say Y here to be able to use netfilter marks as u32 key. + +config NET_CLS_RSVP + tristate "IPv4 Resource Reservation Protocol (RSVP)" + select NET_CLS + ---help--- + The Resource Reservation Protocol (RSVP) permits end systems to + request a minimum and maximum data flow rate for a connection; this + is important for real time data such as streaming sound or video. + + Say Y here if you want to be able to classify outgoing packets based + on their RSVP requests. + + To compile this code as a module, choose M here: the + module will be called cls_rsvp. + +config NET_CLS_RSVP6 + tristate "IPv6 Resource Reservation Protocol (RSVP6)" + select NET_CLS + ---help--- + The Resource Reservation Protocol (RSVP) permits end systems to + request a minimum and maximum data flow rate for a connection; this + is important for real time data such as streaming sound or video. + + Say Y here if you want to be able to classify outgoing packets based + on their RSVP requests and you are using the IPv6 protocol. + + To compile this code as a module, choose M here: the + module will be called cls_rsvp6. + +config NET_CLS_FLOW + tristate "Flow classifier" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets based on + a configurable combination of packet keys. This is mostly useful + in combination with SFQ. + + To compile this code as a module, choose M here: the + module will be called cls_flow. + +config NET_CLS_CGROUP + tristate "Control Group Classifier" + select NET_CLS + select CGROUP_NET_CLASSID + depends on CGROUPS + ---help--- + Say Y here if you want to classify packets based on the control + cgroup of their process. + + To compile this code as a module, choose M here: the + module will be called cls_cgroup. + +config NET_CLS_BPF + tristate "BPF-based classifier" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets based on + programmable BPF (JIT'ed) filters as an alternative to ematches. + + To compile this code as a module, choose M here: the module will + be called cls_bpf. + +config NET_CLS_FLOWER + tristate "Flower classifier" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets based on + a configurable combination of packet keys and masks. + + To compile this code as a module, choose M here: the module will + be called cls_flower. + +config NET_CLS_MATCHALL + tristate "Match-all classifier" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets based on + nothing. Every packet will match. + + To compile this code as a module, choose M here: the module will + be called cls_matchall. + +config NET_EMATCH + bool "Extended Matches" + select NET_CLS + ---help--- + Say Y here if you want to use extended matches on top of classifiers + and select the extended matches below. + + Extended matches are small classification helpers not worth writing + a separate classifier for. + + A recent version of the iproute2 package is required to use + extended matches. + +config NET_EMATCH_STACK + int "Stack size" + depends on NET_EMATCH + default "32" + ---help--- + Size of the local stack variable used while evaluating the tree of + ematches. Limits the depth of the tree, i.e. the number of + encapsulated precedences. Every level requires 4 bytes of additional + stack space. + +config NET_EMATCH_CMP + tristate "Simple packet data comparison" + depends on NET_EMATCH + ---help--- + Say Y here if you want to be able to classify packets based on + simple packet data comparisons for 8, 16, and 32bit values. + + To compile this code as a module, choose M here: the + module will be called em_cmp. + +config NET_EMATCH_NBYTE + tristate "Multi byte comparison" + depends on NET_EMATCH + ---help--- + Say Y here if you want to be able to classify packets based on + multiple byte comparisons mainly useful for IPv6 address comparisons. + + To compile this code as a module, choose M here: the + module will be called em_nbyte. + +config NET_EMATCH_U32 + tristate "U32 key" + depends on NET_EMATCH + ---help--- + Say Y here if you want to be able to classify packets using + the famous u32 key in combination with logic relations. + + To compile this code as a module, choose M here: the + module will be called em_u32. + +config NET_EMATCH_META + tristate "Metadata" + depends on NET_EMATCH + ---help--- + Say Y here if you want to be able to classify packets based on + metadata such as load average, netfilter attributes, socket + attributes and routing decisions. + + To compile this code as a module, choose M here: the + module will be called em_meta. + +config NET_EMATCH_TEXT + tristate "Textsearch" + depends on NET_EMATCH + select TEXTSEARCH + select TEXTSEARCH_KMP + select TEXTSEARCH_BM + select TEXTSEARCH_FSM + ---help--- + Say Y here if you want to be able to classify packets based on + textsearch comparisons. + + To compile this code as a module, choose M here: the + module will be called em_text. + +config NET_EMATCH_CANID + tristate "CAN Identifier" + depends on NET_EMATCH && (CAN=y || CAN=m) + ---help--- + Say Y here if you want to be able to classify CAN frames based + on CAN Identifier. + + To compile this code as a module, choose M here: the + module will be called em_canid. + +config NET_EMATCH_IPSET + tristate "IPset" + depends on NET_EMATCH && IP_SET + ---help--- + Say Y here if you want to be able to classify packets based on + ipset membership. + + To compile this code as a module, choose M here: the + module will be called em_ipset. + +config NET_EMATCH_IPT + tristate "IPtables Matches" + depends on NET_EMATCH && NETFILTER && NETFILTER_XTABLES + ---help--- + Say Y here to be able to classify packets based on iptables + matches. + Current supported match is "policy" which allows packet classification + based on IPsec policy that was used during decapsulation + + To compile this code as a module, choose M here: the + module will be called em_ipt. + +config NET_CLS_ACT + bool "Actions" + select NET_CLS + ---help--- + Say Y here if you want to use traffic control actions. Actions + get attached to classifiers and are invoked after a successful + classification. They are used to overwrite the classification + result, instantly drop or redirect packets, etc. + + A recent version of the iproute2 package is required to use + extended matches. + +config NET_ACT_POLICE + tristate "Traffic Policing" + depends on NET_CLS_ACT + ---help--- + Say Y here if you want to do traffic policing, i.e. strict + bandwidth limiting. This action replaces the existing policing + module. + + To compile this code as a module, choose M here: the + module will be called act_police. + +config NET_ACT_GACT + tristate "Generic actions" + depends on NET_CLS_ACT + ---help--- + Say Y here to take generic actions such as dropping and + accepting packets. + + To compile this code as a module, choose M here: the + module will be called act_gact. + +config GACT_PROB + bool "Probability support" + depends on NET_ACT_GACT + ---help--- + Say Y here to use the generic action randomly or deterministically. + +config NET_ACT_MIRRED + tristate "Redirecting and Mirroring" + depends on NET_CLS_ACT + ---help--- + Say Y here to allow packets to be mirrored or redirected to + other devices. + + To compile this code as a module, choose M here: the + module will be called act_mirred. + +config NET_ACT_SAMPLE + tristate "Traffic Sampling" + depends on NET_CLS_ACT + select PSAMPLE + ---help--- + Say Y here to allow packet sampling tc action. The packet sample + action consists of statistically choosing packets and sampling + them using the psample module. + + To compile this code as a module, choose M here: the + module will be called act_sample. + +config NET_ACT_IPT + tristate "IPtables targets" + depends on NET_CLS_ACT && NETFILTER && IP_NF_IPTABLES + ---help--- + Say Y here to be able to invoke iptables targets after successful + classification. + + To compile this code as a module, choose M here: the + module will be called act_ipt. + +config NET_ACT_NAT + tristate "Stateless NAT" + depends on NET_CLS_ACT + ---help--- + Say Y here to do stateless NAT on IPv4 packets. You should use + netfilter for NAT unless you know what you are doing. + + To compile this code as a module, choose M here: the + module will be called act_nat. + +config NET_ACT_PEDIT + tristate "Packet Editing" + depends on NET_CLS_ACT + ---help--- + Say Y here if you want to mangle the content of packets. + + To compile this code as a module, choose M here: the + module will be called act_pedit. + +config NET_ACT_SIMP + tristate "Simple Example (Debug)" + depends on NET_CLS_ACT + ---help--- + Say Y here to add a simple action for demonstration purposes. + It is meant as an example and for debugging purposes. It will + print a configured policy string followed by the packet count + to the console for every packet that passes by. + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_simple. + +config NET_ACT_SKBEDIT + tristate "SKB Editing" + depends on NET_CLS_ACT + ---help--- + Say Y here to change skb priority or queue_mapping settings. + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_skbedit. + +config NET_ACT_CSUM + tristate "Checksum Updating" + depends on NET_CLS_ACT && INET + select LIBCRC32C + ---help--- + Say Y here to update some common checksum after some direct + packet alterations. + + To compile this code as a module, choose M here: the + module will be called act_csum. + +config NET_ACT_VLAN + tristate "Vlan manipulation" + depends on NET_CLS_ACT + ---help--- + Say Y here to push or pop vlan headers. + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_vlan. + +config NET_ACT_BPF + tristate "BPF based action" + depends on NET_CLS_ACT + ---help--- + Say Y here to execute BPF code on packets. The BPF code will decide + if the packet should be dropped or not. + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_bpf. + +config NET_ACT_CONNMARK + tristate "Netfilter Connection Mark Retriever" + depends on NET_CLS_ACT && NETFILTER && IP_NF_IPTABLES + depends on NF_CONNTRACK && NF_CONNTRACK_MARK + ---help--- + Say Y here to allow retrieving of conn mark + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_connmark. + +config NET_ACT_SKBMOD + tristate "skb data modification action" + depends on NET_CLS_ACT + ---help--- + Say Y here to allow modification of skb data + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_skbmod. + +config NET_ACT_IFE + tristate "Inter-FE action based on IETF ForCES InterFE LFB" + depends on NET_CLS_ACT + select NET_IFE + ---help--- + Say Y here to allow for sourcing and terminating metadata + For details refer to netdev01 paper: + "Distributing Linux Traffic Control Classifier-Action Subsystem" + Authors: Jamal Hadi Salim and Damascene M. Joachimpillai + + To compile this code as a module, choose M here: the + module will be called act_ife. + +config NET_ACT_TUNNEL_KEY + tristate "IP tunnel metadata manipulation" + depends on NET_CLS_ACT + ---help--- + Say Y here to set/release ip tunnel metadata. + + If unsure, say N. + + To compile this code as a module, choose M here: the + module will be called act_tunnel_key. + +config NET_IFE_SKBMARK + tristate "Support to encoding decoding skb mark on IFE action" + depends on NET_ACT_IFE + +config NET_IFE_SKBPRIO + tristate "Support to encoding decoding skb prio on IFE action" + depends on NET_ACT_IFE + +config NET_IFE_SKBTCINDEX + tristate "Support to encoding decoding skb tcindex on IFE action" + depends on NET_ACT_IFE + +config NET_CLS_IND + bool "Incoming device classification" + depends on NET_CLS_U32 || NET_CLS_FW + ---help--- + Say Y here to extend the u32 and fw classifier to support + classification based on the incoming device. This option is + likely to disappear in favour of the metadata ematch. + +endif # NET_SCHED + +config NET_SCH_FIFO + bool diff --git a/net/sched/Makefile b/net/sched/Makefile new file mode 100644 index 000000000..f0403f49e --- /dev/null +++ b/net/sched/Makefile @@ -0,0 +1,81 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# Makefile for the Linux Traffic Control Unit. +# + +obj-y := sch_generic.o sch_mq.o + +obj-$(CONFIG_NET_SCHED) += sch_api.o sch_blackhole.o +obj-$(CONFIG_NET_CLS) += cls_api.o +obj-$(CONFIG_NET_CLS_ACT) += act_api.o +obj-$(CONFIG_NET_ACT_POLICE) += act_police.o +obj-$(CONFIG_NET_ACT_GACT) += act_gact.o +obj-$(CONFIG_NET_ACT_MIRRED) += act_mirred.o +obj-$(CONFIG_NET_ACT_SAMPLE) += act_sample.o +obj-$(CONFIG_NET_ACT_IPT) += act_ipt.o +obj-$(CONFIG_NET_ACT_NAT) += act_nat.o +obj-$(CONFIG_NET_ACT_PEDIT) += act_pedit.o +obj-$(CONFIG_NET_ACT_SIMP) += act_simple.o +obj-$(CONFIG_NET_ACT_SKBEDIT) += act_skbedit.o +obj-$(CONFIG_NET_ACT_CSUM) += act_csum.o +obj-$(CONFIG_NET_ACT_VLAN) += act_vlan.o +obj-$(CONFIG_NET_ACT_BPF) += act_bpf.o +obj-$(CONFIG_NET_ACT_CONNMARK) += act_connmark.o +obj-$(CONFIG_NET_ACT_SKBMOD) += act_skbmod.o +obj-$(CONFIG_NET_ACT_IFE) += act_ife.o +obj-$(CONFIG_NET_IFE_SKBMARK) += act_meta_mark.o +obj-$(CONFIG_NET_IFE_SKBPRIO) += act_meta_skbprio.o +obj-$(CONFIG_NET_IFE_SKBTCINDEX) += act_meta_skbtcindex.o +obj-$(CONFIG_NET_ACT_TUNNEL_KEY)+= act_tunnel_key.o +obj-$(CONFIG_NET_SCH_FIFO) += sch_fifo.o +obj-$(CONFIG_NET_SCH_CBQ) += sch_cbq.o +obj-$(CONFIG_NET_SCH_HTB) += sch_htb.o +obj-$(CONFIG_NET_SCH_HFSC) += sch_hfsc.o +obj-$(CONFIG_NET_SCH_RED) += sch_red.o +obj-$(CONFIG_NET_SCH_GRED) += sch_gred.o +obj-$(CONFIG_NET_SCH_INGRESS) += sch_ingress.o +obj-$(CONFIG_NET_SCH_DSMARK) += sch_dsmark.o +obj-$(CONFIG_NET_SCH_SFB) += sch_sfb.o +obj-$(CONFIG_NET_SCH_SFQ) += sch_sfq.o +obj-$(CONFIG_NET_SCH_TBF) += sch_tbf.o +obj-$(CONFIG_NET_SCH_TEQL) += sch_teql.o +obj-$(CONFIG_NET_SCH_PRIO) += sch_prio.o +obj-$(CONFIG_NET_SCH_MULTIQ) += sch_multiq.o +obj-$(CONFIG_NET_SCH_ATM) += sch_atm.o +obj-$(CONFIG_NET_SCH_NETEM) += sch_netem.o +obj-$(CONFIG_NET_SCH_DRR) += sch_drr.o +obj-$(CONFIG_NET_SCH_PLUG) += sch_plug.o +obj-$(CONFIG_NET_SCH_MQPRIO) += sch_mqprio.o +obj-$(CONFIG_NET_SCH_SKBPRIO) += sch_skbprio.o +obj-$(CONFIG_NET_SCH_CHOKE) += sch_choke.o +obj-$(CONFIG_NET_SCH_QFQ) += sch_qfq.o +obj-$(CONFIG_NET_SCH_CODEL) += sch_codel.o +obj-$(CONFIG_NET_SCH_FQ_CODEL) += sch_fq_codel.o +obj-$(CONFIG_NET_SCH_CAKE) += sch_cake.o +obj-$(CONFIG_NET_SCH_FQ) += sch_fq.o +obj-$(CONFIG_NET_SCH_HHF) += sch_hhf.o +obj-$(CONFIG_NET_SCH_PIE) += sch_pie.o +obj-$(CONFIG_NET_SCH_CBS) += sch_cbs.o +obj-$(CONFIG_NET_SCH_ETF) += sch_etf.o + +obj-$(CONFIG_NET_CLS_U32) += cls_u32.o +obj-$(CONFIG_NET_CLS_ROUTE4) += cls_route.o +obj-$(CONFIG_NET_CLS_FW) += cls_fw.o +obj-$(CONFIG_NET_CLS_RSVP) += cls_rsvp.o +obj-$(CONFIG_NET_CLS_TCINDEX) += cls_tcindex.o +obj-$(CONFIG_NET_CLS_RSVP6) += cls_rsvp6.o +obj-$(CONFIG_NET_CLS_BASIC) += cls_basic.o +obj-$(CONFIG_NET_CLS_FLOW) += cls_flow.o +obj-$(CONFIG_NET_CLS_CGROUP) += cls_cgroup.o +obj-$(CONFIG_NET_CLS_BPF) += cls_bpf.o +obj-$(CONFIG_NET_CLS_FLOWER) += cls_flower.o +obj-$(CONFIG_NET_CLS_MATCHALL) += cls_matchall.o +obj-$(CONFIG_NET_EMATCH) += ematch.o +obj-$(CONFIG_NET_EMATCH_CMP) += em_cmp.o +obj-$(CONFIG_NET_EMATCH_NBYTE) += em_nbyte.o +obj-$(CONFIG_NET_EMATCH_U32) += em_u32.o +obj-$(CONFIG_NET_EMATCH_META) += em_meta.o +obj-$(CONFIG_NET_EMATCH_TEXT) += em_text.o +obj-$(CONFIG_NET_EMATCH_CANID) += em_canid.o +obj-$(CONFIG_NET_EMATCH_IPSET) += em_ipset.o +obj-$(CONFIG_NET_EMATCH_IPT) += em_ipt.o diff --git a/net/sched/act_api.c b/net/sched/act_api.c new file mode 100644 index 000000000..ad0773b20 --- /dev/null +++ b/net/sched/act_api.c @@ -0,0 +1,1739 @@ +/* + * net/sched/act_api.c Packet action API. + * + * 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. + * + * Author: Jamal Hadi Salim + * + * + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/skbuff.h> +#include <linux/init.h> +#include <linux/kmod.h> +#include <linux/err.h> +#include <linux/module.h> +#include <linux/rhashtable.h> +#include <linux/list.h> +#include <net/net_namespace.h> +#include <net/sock.h> +#include <net/sch_generic.h> +#include <net/pkt_cls.h> +#include <net/act_api.h> +#include <net/netlink.h> + +static int tcf_action_goto_chain_init(struct tc_action *a, struct tcf_proto *tp) +{ + u32 chain_index = a->tcfa_action & TC_ACT_EXT_VAL_MASK; + + if (!tp) + return -EINVAL; + a->goto_chain = tcf_chain_get_by_act(tp->chain->block, chain_index); + if (!a->goto_chain) + return -ENOMEM; + return 0; +} + +static void tcf_action_goto_chain_fini(struct tc_action *a) +{ + tcf_chain_put_by_act(a->goto_chain); +} + +static void tcf_action_goto_chain_exec(const struct tc_action *a, + struct tcf_result *res) +{ + const struct tcf_chain *chain = a->goto_chain; + + res->goto_tp = rcu_dereference_bh(chain->filter_chain); +} + +static void tcf_free_cookie_rcu(struct rcu_head *p) +{ + struct tc_cookie *cookie = container_of(p, struct tc_cookie, rcu); + + kfree(cookie->data); + kfree(cookie); +} + +static void tcf_set_action_cookie(struct tc_cookie __rcu **old_cookie, + struct tc_cookie *new_cookie) +{ + struct tc_cookie *old; + + old = xchg((__force struct tc_cookie **)old_cookie, new_cookie); + if (old) + call_rcu(&old->rcu, tcf_free_cookie_rcu); +} + +/* XXX: For standalone actions, we don't need a RCU grace period either, because + * actions are always connected to filters and filters are already destroyed in + * RCU callbacks, so after a RCU grace period actions are already disconnected + * from filters. Readers later can not find us. + */ +static void free_tcf(struct tc_action *p) +{ + free_percpu(p->cpu_bstats); + free_percpu(p->cpu_qstats); + + tcf_set_action_cookie(&p->act_cookie, NULL); + if (p->goto_chain) + tcf_action_goto_chain_fini(p); + + kfree(p); +} + +static void tcf_action_cleanup(struct tc_action *p) +{ + if (p->ops->cleanup) + p->ops->cleanup(p); + + gen_kill_estimator(&p->tcfa_rate_est); + free_tcf(p); +} + +static int __tcf_action_put(struct tc_action *p, bool bind) +{ + struct tcf_idrinfo *idrinfo = p->idrinfo; + + if (refcount_dec_and_lock(&p->tcfa_refcnt, &idrinfo->lock)) { + if (bind) + atomic_dec(&p->tcfa_bindcnt); + idr_remove(&idrinfo->action_idr, p->tcfa_index); + spin_unlock(&idrinfo->lock); + + tcf_action_cleanup(p); + return 1; + } + + if (bind) + atomic_dec(&p->tcfa_bindcnt); + + return 0; +} + +int __tcf_idr_release(struct tc_action *p, bool bind, bool strict) +{ + int ret = 0; + + /* Release with strict==1 and bind==0 is only called through act API + * interface (classifiers always bind). Only case when action with + * positive reference count and zero bind count can exist is when it was + * also created with act API (unbinding last classifier will destroy the + * action if it was created by classifier). So only case when bind count + * can be changed after initial check is when unbound action is + * destroyed by act API while classifier binds to action with same id + * concurrently. This result either creation of new action(same behavior + * as before), or reusing existing action if concurrent process + * increments reference count before action is deleted. Both scenarios + * are acceptable. + */ + if (p) { + if (!bind && strict && atomic_read(&p->tcfa_bindcnt) > 0) + return -EPERM; + + if (__tcf_action_put(p, bind)) + ret = ACT_P_DELETED; + } + + return ret; +} +EXPORT_SYMBOL(__tcf_idr_release); + +static size_t tcf_action_shared_attrs_size(const struct tc_action *act) +{ + struct tc_cookie *act_cookie; + u32 cookie_len = 0; + + rcu_read_lock(); + act_cookie = rcu_dereference(act->act_cookie); + + if (act_cookie) + cookie_len = nla_total_size(act_cookie->len); + rcu_read_unlock(); + + return nla_total_size(0) /* action number nested */ + + nla_total_size(IFNAMSIZ) /* TCA_ACT_KIND */ + + cookie_len /* TCA_ACT_COOKIE */ + + nla_total_size(0) /* TCA_ACT_STATS nested */ + /* TCA_STATS_BASIC */ + + nla_total_size_64bit(sizeof(struct gnet_stats_basic)) + /* TCA_STATS_QUEUE */ + + nla_total_size_64bit(sizeof(struct gnet_stats_queue)) + + nla_total_size(0) /* TCA_OPTIONS nested */ + + nla_total_size(sizeof(struct tcf_t)); /* TCA_GACT_TM */ +} + +static size_t tcf_action_full_attrs_size(size_t sz) +{ + return NLMSG_HDRLEN /* struct nlmsghdr */ + + sizeof(struct tcamsg) + + nla_total_size(0) /* TCA_ACT_TAB nested */ + + sz; +} + +static size_t tcf_action_fill_size(const struct tc_action *act) +{ + size_t sz = tcf_action_shared_attrs_size(act); + + if (act->ops->get_fill_size) + return act->ops->get_fill_size(act) + sz; + return sz; +} + +static int tcf_dump_walker(struct tcf_idrinfo *idrinfo, struct sk_buff *skb, + struct netlink_callback *cb) +{ + int err = 0, index = -1, s_i = 0, n_i = 0; + u32 act_flags = cb->args[2]; + unsigned long jiffy_since = cb->args[3]; + struct nlattr *nest; + struct idr *idr = &idrinfo->action_idr; + struct tc_action *p; + unsigned long id = 1; + + spin_lock(&idrinfo->lock); + + s_i = cb->args[0]; + + idr_for_each_entry_ul(idr, p, id) { + index++; + if (index < s_i) + continue; + + if (jiffy_since && + time_after(jiffy_since, + (unsigned long)p->tcfa_tm.lastuse)) + continue; + + nest = nla_nest_start(skb, n_i); + if (!nest) { + index--; + goto nla_put_failure; + } + err = tcf_action_dump_1(skb, p, 0, 0); + if (err < 0) { + index--; + nlmsg_trim(skb, nest); + goto done; + } + nla_nest_end(skb, nest); + n_i++; + if (!(act_flags & TCA_FLAG_LARGE_DUMP_ON) && + n_i >= TCA_ACT_MAX_PRIO) + goto done; + } +done: + if (index >= 0) + cb->args[0] = index + 1; + + spin_unlock(&idrinfo->lock); + if (n_i) { + if (act_flags & TCA_FLAG_LARGE_DUMP_ON) + cb->args[1] = n_i; + } + return n_i; + +nla_put_failure: + nla_nest_cancel(skb, nest); + goto done; +} + +static int tcf_del_walker(struct tcf_idrinfo *idrinfo, struct sk_buff *skb, + const struct tc_action_ops *ops) +{ + struct nlattr *nest; + int n_i = 0; + int ret = -EINVAL; + struct idr *idr = &idrinfo->action_idr; + struct tc_action *p; + unsigned long id = 1; + + nest = nla_nest_start(skb, 0); + if (nest == NULL) + goto nla_put_failure; + if (nla_put_string(skb, TCA_KIND, ops->kind)) + goto nla_put_failure; + + idr_for_each_entry_ul(idr, p, id) { + ret = __tcf_idr_release(p, false, true); + if (ret == ACT_P_DELETED) { + module_put(ops->owner); + n_i++; + } else if (ret < 0) { + goto nla_put_failure; + } + } + if (nla_put_u32(skb, TCA_FCNT, n_i)) + goto nla_put_failure; + nla_nest_end(skb, nest); + + return n_i; +nla_put_failure: + nla_nest_cancel(skb, nest); + return ret; +} + +int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tcf_idrinfo *idrinfo = tn->idrinfo; + + if (type == RTM_DELACTION) { + return tcf_del_walker(idrinfo, skb, ops); + } else if (type == RTM_GETACTION) { + return tcf_dump_walker(idrinfo, skb, cb); + } else { + WARN(1, "tcf_generic_walker: unknown command %d\n", type); + NL_SET_ERR_MSG(extack, "tcf_generic_walker: unknown command"); + return -EINVAL; + } +} +EXPORT_SYMBOL(tcf_generic_walker); + +int tcf_idr_search(struct tc_action_net *tn, struct tc_action **a, u32 index) +{ + struct tcf_idrinfo *idrinfo = tn->idrinfo; + struct tc_action *p; + + spin_lock(&idrinfo->lock); + p = idr_find(&idrinfo->action_idr, index); + if (IS_ERR(p)) + p = NULL; + else if (p) + refcount_inc(&p->tcfa_refcnt); + spin_unlock(&idrinfo->lock); + + if (p) { + *a = p; + return true; + } + return false; +} +EXPORT_SYMBOL(tcf_idr_search); + +static int tcf_idr_delete_index(struct tcf_idrinfo *idrinfo, u32 index) +{ + struct tc_action *p; + int ret = 0; + + spin_lock(&idrinfo->lock); + p = idr_find(&idrinfo->action_idr, index); + if (!p) { + spin_unlock(&idrinfo->lock); + return -ENOENT; + } + + if (!atomic_read(&p->tcfa_bindcnt)) { + if (refcount_dec_and_test(&p->tcfa_refcnt)) { + struct module *owner = p->ops->owner; + + WARN_ON(p != idr_remove(&idrinfo->action_idr, + p->tcfa_index)); + spin_unlock(&idrinfo->lock); + + tcf_action_cleanup(p); + module_put(owner); + return 0; + } + ret = 0; + } else { + ret = -EPERM; + } + + spin_unlock(&idrinfo->lock); + return ret; +} + +int tcf_idr_create(struct tc_action_net *tn, u32 index, struct nlattr *est, + struct tc_action **a, const struct tc_action_ops *ops, + int bind, bool cpustats) +{ + struct tc_action *p = kzalloc(ops->size, GFP_KERNEL); + struct tcf_idrinfo *idrinfo = tn->idrinfo; + int err = -ENOMEM; + + if (unlikely(!p)) + return -ENOMEM; + refcount_set(&p->tcfa_refcnt, 1); + if (bind) + atomic_set(&p->tcfa_bindcnt, 1); + + if (cpustats) { + p->cpu_bstats = netdev_alloc_pcpu_stats(struct gnet_stats_basic_cpu); + if (!p->cpu_bstats) + goto err1; + p->cpu_qstats = alloc_percpu(struct gnet_stats_queue); + if (!p->cpu_qstats) + goto err2; + } + spin_lock_init(&p->tcfa_lock); + p->tcfa_index = index; + p->tcfa_tm.install = jiffies; + p->tcfa_tm.lastuse = jiffies; + p->tcfa_tm.firstuse = 0; + if (est) { + err = gen_new_estimator(&p->tcfa_bstats, p->cpu_bstats, + &p->tcfa_rate_est, + &p->tcfa_lock, NULL, est); + if (err) + goto err3; + } + + p->idrinfo = idrinfo; + p->ops = ops; + *a = p; + return 0; +err3: + free_percpu(p->cpu_qstats); +err2: + free_percpu(p->cpu_bstats); +err1: + kfree(p); + return err; +} +EXPORT_SYMBOL(tcf_idr_create); + +void tcf_idr_insert(struct tc_action_net *tn, struct tc_action *a) +{ + struct tcf_idrinfo *idrinfo = tn->idrinfo; + + spin_lock(&idrinfo->lock); + /* Replace ERR_PTR(-EBUSY) allocated by tcf_idr_check_alloc */ + WARN_ON(!IS_ERR(idr_replace(&idrinfo->action_idr, a, a->tcfa_index))); + spin_unlock(&idrinfo->lock); +} +EXPORT_SYMBOL(tcf_idr_insert); + +/* Cleanup idr index that was allocated but not initialized. */ + +void tcf_idr_cleanup(struct tc_action_net *tn, u32 index) +{ + struct tcf_idrinfo *idrinfo = tn->idrinfo; + + spin_lock(&idrinfo->lock); + /* Remove ERR_PTR(-EBUSY) allocated by tcf_idr_check_alloc */ + WARN_ON(!IS_ERR(idr_remove(&idrinfo->action_idr, index))); + spin_unlock(&idrinfo->lock); +} +EXPORT_SYMBOL(tcf_idr_cleanup); + +/* Check if action with specified index exists. If actions is found, increments + * its reference and bind counters, and return 1. Otherwise insert temporary + * error pointer (to prevent concurrent users from inserting actions with same + * index) and return 0. + */ + +int tcf_idr_check_alloc(struct tc_action_net *tn, u32 *index, + struct tc_action **a, int bind) +{ + struct tcf_idrinfo *idrinfo = tn->idrinfo; + struct tc_action *p; + int ret; + +again: + spin_lock(&idrinfo->lock); + if (*index) { + p = idr_find(&idrinfo->action_idr, *index); + if (IS_ERR(p)) { + /* This means that another process allocated + * index but did not assign the pointer yet. + */ + spin_unlock(&idrinfo->lock); + goto again; + } + + if (p) { + refcount_inc(&p->tcfa_refcnt); + if (bind) + atomic_inc(&p->tcfa_bindcnt); + *a = p; + ret = 1; + } else { + *a = NULL; + ret = idr_alloc_u32(&idrinfo->action_idr, NULL, index, + *index, GFP_ATOMIC); + if (!ret) + idr_replace(&idrinfo->action_idr, + ERR_PTR(-EBUSY), *index); + } + } else { + *index = 1; + *a = NULL; + ret = idr_alloc_u32(&idrinfo->action_idr, NULL, index, + UINT_MAX, GFP_ATOMIC); + if (!ret) + idr_replace(&idrinfo->action_idr, ERR_PTR(-EBUSY), + *index); + } + spin_unlock(&idrinfo->lock); + return ret; +} +EXPORT_SYMBOL(tcf_idr_check_alloc); + +void tcf_idrinfo_destroy(const struct tc_action_ops *ops, + struct tcf_idrinfo *idrinfo) +{ + struct idr *idr = &idrinfo->action_idr; + struct tc_action *p; + int ret; + unsigned long id = 1; + + idr_for_each_entry_ul(idr, p, id) { + ret = __tcf_idr_release(p, false, true); + if (ret == ACT_P_DELETED) + module_put(ops->owner); + else if (ret < 0) + return; + } + idr_destroy(&idrinfo->action_idr); +} +EXPORT_SYMBOL(tcf_idrinfo_destroy); + +static LIST_HEAD(act_base); +static DEFINE_RWLOCK(act_mod_lock); + +int tcf_register_action(struct tc_action_ops *act, + struct pernet_operations *ops) +{ + struct tc_action_ops *a; + int ret; + + if (!act->act || !act->dump || !act->init || !act->walk || !act->lookup) + return -EINVAL; + + /* We have to register pernet ops before making the action ops visible, + * otherwise tcf_action_init_1() could get a partially initialized + * netns. + */ + ret = register_pernet_subsys(ops); + if (ret) + return ret; + + write_lock(&act_mod_lock); + list_for_each_entry(a, &act_base, head) { + if (act->type == a->type || (strcmp(act->kind, a->kind) == 0)) { + write_unlock(&act_mod_lock); + unregister_pernet_subsys(ops); + return -EEXIST; + } + } + list_add_tail(&act->head, &act_base); + write_unlock(&act_mod_lock); + + return 0; +} +EXPORT_SYMBOL(tcf_register_action); + +int tcf_unregister_action(struct tc_action_ops *act, + struct pernet_operations *ops) +{ + struct tc_action_ops *a; + int err = -ENOENT; + + write_lock(&act_mod_lock); + list_for_each_entry(a, &act_base, head) { + if (a == act) { + list_del(&act->head); + err = 0; + break; + } + } + write_unlock(&act_mod_lock); + if (!err) + unregister_pernet_subsys(ops); + return err; +} +EXPORT_SYMBOL(tcf_unregister_action); + +/* lookup by name */ +static struct tc_action_ops *tc_lookup_action_n(char *kind) +{ + struct tc_action_ops *a, *res = NULL; + + if (kind) { + read_lock(&act_mod_lock); + list_for_each_entry(a, &act_base, head) { + if (strcmp(kind, a->kind) == 0) { + if (try_module_get(a->owner)) + res = a; + break; + } + } + read_unlock(&act_mod_lock); + } + return res; +} + +/* lookup by nlattr */ +static struct tc_action_ops *tc_lookup_action(struct nlattr *kind) +{ + struct tc_action_ops *a, *res = NULL; + + if (kind) { + read_lock(&act_mod_lock); + list_for_each_entry(a, &act_base, head) { + if (nla_strcmp(kind, a->kind) == 0) { + if (try_module_get(a->owner)) + res = a; + break; + } + } + read_unlock(&act_mod_lock); + } + return res; +} + +/*TCA_ACT_MAX_PRIO is 32, there count upto 32 */ +#define TCA_ACT_MAX_PRIO_MASK 0x1FF +int tcf_action_exec(struct sk_buff *skb, struct tc_action **actions, + int nr_actions, struct tcf_result *res) +{ + u32 jmp_prgcnt = 0; + u32 jmp_ttl = TCA_ACT_MAX_PRIO; /*matches actions per filter */ + int i; + int ret = TC_ACT_OK; + + if (skb_skip_tc_classify(skb)) + return TC_ACT_OK; + +restart_act_graph: + for (i = 0; i < nr_actions; i++) { + const struct tc_action *a = actions[i]; + int repeat_ttl; + + if (jmp_prgcnt > 0) { + jmp_prgcnt -= 1; + continue; + } + + repeat_ttl = 32; +repeat: + ret = a->ops->act(skb, a, res); + + if (unlikely(ret == TC_ACT_REPEAT)) { + if (--repeat_ttl != 0) + goto repeat; + /* suspicious opcode, stop pipeline */ + net_warn_ratelimited("TC_ACT_REPEAT abuse ?\n"); + return TC_ACT_OK; + } + + if (TC_ACT_EXT_CMP(ret, TC_ACT_JUMP)) { + jmp_prgcnt = ret & TCA_ACT_MAX_PRIO_MASK; + if (!jmp_prgcnt || (jmp_prgcnt > nr_actions)) { + /* faulty opcode, stop pipeline */ + return TC_ACT_OK; + } else { + jmp_ttl -= 1; + if (jmp_ttl > 0) + goto restart_act_graph; + else /* faulty graph, stop pipeline */ + return TC_ACT_OK; + } + } else if (TC_ACT_EXT_CMP(ret, TC_ACT_GOTO_CHAIN)) { + tcf_action_goto_chain_exec(a, res); + } + + if (ret != TC_ACT_PIPE) + break; + } + + return ret; +} +EXPORT_SYMBOL(tcf_action_exec); + +int tcf_action_destroy(struct tc_action *actions[], int bind) +{ + const struct tc_action_ops *ops; + struct tc_action *a; + int ret = 0, i; + + for (i = 0; i < TCA_ACT_MAX_PRIO && actions[i]; i++) { + a = actions[i]; + actions[i] = NULL; + ops = a->ops; + ret = __tcf_idr_release(a, bind, true); + if (ret == ACT_P_DELETED) + module_put(ops->owner); + else if (ret < 0) + return ret; + } + return ret; +} + +static int tcf_action_destroy_1(struct tc_action *a, int bind) +{ + struct tc_action *actions[] = { a, NULL }; + + return tcf_action_destroy(actions, bind); +} + +static int tcf_action_put(struct tc_action *p) +{ + return __tcf_action_put(p, false); +} + +/* Put all actions in this array, skip those NULL's. */ +static void tcf_action_put_many(struct tc_action *actions[]) +{ + int i; + + for (i = 0; i < TCA_ACT_MAX_PRIO; i++) { + struct tc_action *a = actions[i]; + const struct tc_action_ops *ops; + + if (!a) + continue; + ops = a->ops; + if (tcf_action_put(a)) + module_put(ops->owner); + } +} + +int +tcf_action_dump_old(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +{ + return a->ops->dump(skb, a, bind, ref); +} + +int +tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +{ + int err = -EINVAL; + unsigned char *b = skb_tail_pointer(skb); + struct nlattr *nest; + struct tc_cookie *cookie; + + if (nla_put_string(skb, TCA_KIND, a->ops->kind)) + goto nla_put_failure; + if (tcf_action_copy_stats(skb, a, 0)) + goto nla_put_failure; + + rcu_read_lock(); + cookie = rcu_dereference(a->act_cookie); + if (cookie) { + if (nla_put(skb, TCA_ACT_COOKIE, cookie->len, cookie->data)) { + rcu_read_unlock(); + goto nla_put_failure; + } + } + rcu_read_unlock(); + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + err = tcf_action_dump_old(skb, a, bind, ref); + if (err > 0) { + nla_nest_end(skb, nest); + return err; + } + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} +EXPORT_SYMBOL(tcf_action_dump_1); + +int tcf_action_dump(struct sk_buff *skb, struct tc_action *actions[], + int bind, int ref) +{ + struct tc_action *a; + int err = -EINVAL, i; + struct nlattr *nest; + + for (i = 0; i < TCA_ACT_MAX_PRIO && actions[i]; i++) { + a = actions[i]; + nest = nla_nest_start(skb, i + 1); + if (nest == NULL) + goto nla_put_failure; + err = tcf_action_dump_1(skb, a, bind, ref); + if (err < 0) + goto errout; + nla_nest_end(skb, nest); + } + + return 0; + +nla_put_failure: + err = -EINVAL; +errout: + nla_nest_cancel(skb, nest); + return err; +} + +static struct tc_cookie *nla_memdup_cookie(struct nlattr **tb) +{ + struct tc_cookie *c = kzalloc(sizeof(*c), GFP_KERNEL); + if (!c) + return NULL; + + c->data = nla_memdup(tb[TCA_ACT_COOKIE], GFP_KERNEL); + if (!c->data) { + kfree(c); + return NULL; + } + c->len = nla_len(tb[TCA_ACT_COOKIE]); + + return c; +} + +static bool tcf_action_valid(int action) +{ + int opcode = TC_ACT_EXT_OPCODE(action); + + if (!opcode) + return action <= TC_ACT_VALUE_MAX; + return opcode <= TC_ACT_EXT_OPCODE_MAX || action == TC_ACT_UNSPEC; +} + +struct tc_action *tcf_action_init_1(struct net *net, struct tcf_proto *tp, + struct nlattr *nla, struct nlattr *est, + char *name, int ovr, int bind, + bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action *a; + struct tc_action_ops *a_o; + struct tc_cookie *cookie = NULL; + char act_name[IFNAMSIZ]; + struct nlattr *tb[TCA_ACT_MAX + 1]; + struct nlattr *kind; + int err; + + if (name == NULL) { + err = nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL, extack); + if (err < 0) + goto err_out; + err = -EINVAL; + kind = tb[TCA_ACT_KIND]; + if (!kind) { + NL_SET_ERR_MSG(extack, "TC action kind must be specified"); + goto err_out; + } + if (nla_strlcpy(act_name, kind, IFNAMSIZ) >= IFNAMSIZ) { + NL_SET_ERR_MSG(extack, "TC action name too long"); + goto err_out; + } + if (tb[TCA_ACT_COOKIE]) { + int cklen = nla_len(tb[TCA_ACT_COOKIE]); + + if (cklen > TC_COOKIE_MAX_SIZE) { + NL_SET_ERR_MSG(extack, "TC cookie size above the maximum"); + goto err_out; + } + + cookie = nla_memdup_cookie(tb); + if (!cookie) { + NL_SET_ERR_MSG(extack, "No memory to generate TC cookie"); + err = -ENOMEM; + goto err_out; + } + } + } else { + if (strlcpy(act_name, name, IFNAMSIZ) >= IFNAMSIZ) { + NL_SET_ERR_MSG(extack, "TC action name too long"); + err = -EINVAL; + goto err_out; + } + } + + a_o = tc_lookup_action_n(act_name); + if (a_o == NULL) { +#ifdef CONFIG_MODULES + if (rtnl_held) + rtnl_unlock(); + request_module("act_%s", act_name); + if (rtnl_held) + rtnl_lock(); + + a_o = tc_lookup_action_n(act_name); + + /* We dropped the RTNL semaphore in order to + * perform the module load. So, even if we + * succeeded in loading the module we have to + * tell the caller to replay the request. We + * indicate this using -EAGAIN. + */ + if (a_o != NULL) { + err = -EAGAIN; + goto err_mod; + } +#endif + NL_SET_ERR_MSG(extack, "Failed to load TC action module"); + err = -ENOENT; + goto err_out; + } + + /* backward compatibility for policer */ + if (name == NULL) + err = a_o->init(net, tb[TCA_ACT_OPTIONS], est, &a, ovr, bind, + rtnl_held, extack); + else + err = a_o->init(net, nla, est, &a, ovr, bind, rtnl_held, + extack); + if (err < 0) + goto err_mod; + + if (!name && tb[TCA_ACT_COOKIE]) + tcf_set_action_cookie(&a->act_cookie, cookie); + + /* module count goes up only when brand new policy is created + * if it exists and is only bound to in a_o->init() then + * ACT_P_CREATED is not returned (a zero is). + */ + if (err != ACT_P_CREATED) + module_put(a_o->owner); + + if (TC_ACT_EXT_CMP(a->tcfa_action, TC_ACT_GOTO_CHAIN)) { + err = tcf_action_goto_chain_init(a, tp); + if (err) { + tcf_action_destroy_1(a, bind); + NL_SET_ERR_MSG(extack, "Failed to init TC action chain"); + return ERR_PTR(err); + } + } + + if (!tcf_action_valid(a->tcfa_action)) { + tcf_action_destroy_1(a, bind); + NL_SET_ERR_MSG(extack, "Invalid control action value"); + return ERR_PTR(-EINVAL); + } + + if (!bind && ovr && err == ACT_P_CREATED) + refcount_set(&a->tcfa_refcnt, 2); + + return a; + +err_mod: + module_put(a_o->owner); +err_out: + if (cookie) { + kfree(cookie->data); + kfree(cookie); + } + return ERR_PTR(err); +} + +/* Returns numbers of initialized actions or negative error. */ + +int tcf_action_init(struct net *net, struct tcf_proto *tp, struct nlattr *nla, + struct nlattr *est, char *name, int ovr, int bind, + struct tc_action *actions[], size_t *attr_size, + bool rtnl_held, struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_ACT_MAX_PRIO + 1]; + struct tc_action *act; + size_t sz = 0; + int err; + int i; + + err = nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL, extack); + if (err < 0) + return err; + + for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) { + act = tcf_action_init_1(net, tp, tb[i], est, name, ovr, bind, + rtnl_held, extack); + if (IS_ERR(act)) { + err = PTR_ERR(act); + goto err; + } + act->order = i; + sz += tcf_action_fill_size(act); + /* Start from index 0 */ + actions[i - 1] = act; + } + + *attr_size = tcf_action_full_attrs_size(sz); + return i - 1; + +err: + tcf_action_destroy(actions, bind); + return err; +} + +int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *p, + int compat_mode) +{ + int err = 0; + struct gnet_dump d; + + if (p == NULL) + goto errout; + + /* compat_mode being true specifies a call that is supposed + * to add additional backward compatibility statistic TLVs. + */ + if (compat_mode) { + if (p->type == TCA_OLD_COMPAT) + err = gnet_stats_start_copy_compat(skb, 0, + TCA_STATS, + TCA_XSTATS, + &p->tcfa_lock, &d, + TCA_PAD); + else + return 0; + } else + err = gnet_stats_start_copy(skb, TCA_ACT_STATS, + &p->tcfa_lock, &d, TCA_ACT_PAD); + + if (err < 0) + goto errout; + + if (gnet_stats_copy_basic(NULL, &d, p->cpu_bstats, &p->tcfa_bstats) < 0 || + gnet_stats_copy_rate_est(&d, &p->tcfa_rate_est) < 0 || + gnet_stats_copy_queue(&d, p->cpu_qstats, + &p->tcfa_qstats, + p->tcfa_qstats.qlen) < 0) + goto errout; + + if (gnet_stats_finish_copy(&d) < 0) + goto errout; + + return 0; + +errout: + return -1; +} + +static int tca_get_fill(struct sk_buff *skb, struct tc_action *actions[], + u32 portid, u32 seq, u16 flags, int event, int bind, + int ref) +{ + struct tcamsg *t; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + struct nlattr *nest; + + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*t), flags); + if (!nlh) + goto out_nlmsg_trim; + t = nlmsg_data(nlh); + t->tca_family = AF_UNSPEC; + t->tca__pad1 = 0; + t->tca__pad2 = 0; + + nest = nla_nest_start(skb, TCA_ACT_TAB); + if (!nest) + goto out_nlmsg_trim; + + if (tcf_action_dump(skb, actions, bind, ref) < 0) + goto out_nlmsg_trim; + + nla_nest_end(skb, nest); + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: + nlmsg_trim(skb, b); + return -1; +} + +static int +tcf_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, + struct tc_action *actions[], int event, + struct netlink_ext_ack *extack) +{ + struct sk_buff *skb; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, event, + 0, 1) <= 0) { + NL_SET_ERR_MSG(extack, "Failed to fill netlink attributes while adding TC action"); + kfree_skb(skb); + return -EINVAL; + } + + return rtnl_unicast(skb, net, portid); +} + +static struct tc_action *tcf_action_get_1(struct net *net, struct nlattr *nla, + struct nlmsghdr *n, u32 portid, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_ACT_MAX + 1]; + const struct tc_action_ops *ops; + struct tc_action *a; + int index; + int err; + + err = nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL, extack); + if (err < 0) + goto err_out; + + err = -EINVAL; + if (tb[TCA_ACT_INDEX] == NULL || + nla_len(tb[TCA_ACT_INDEX]) < sizeof(index)) { + NL_SET_ERR_MSG(extack, "Invalid TC action index value"); + goto err_out; + } + index = nla_get_u32(tb[TCA_ACT_INDEX]); + + err = -EINVAL; + ops = tc_lookup_action(tb[TCA_ACT_KIND]); + if (!ops) { /* could happen in batch of actions */ + NL_SET_ERR_MSG(extack, "Specified TC action not found"); + goto err_out; + } + err = -ENOENT; + if (ops->lookup(net, &a, index, extack) == 0) + goto err_mod; + + module_put(ops->owner); + return a; + +err_mod: + module_put(ops->owner); +err_out: + return ERR_PTR(err); +} + +static int tca_action_flush(struct net *net, struct nlattr *nla, + struct nlmsghdr *n, u32 portid, + struct netlink_ext_ack *extack) +{ + struct sk_buff *skb; + unsigned char *b; + struct nlmsghdr *nlh; + struct tcamsg *t; + struct netlink_callback dcb; + struct nlattr *nest; + struct nlattr *tb[TCA_ACT_MAX + 1]; + const struct tc_action_ops *ops; + struct nlattr *kind; + int err = -ENOMEM; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return err; + + b = skb_tail_pointer(skb); + + err = nla_parse_nested(tb, TCA_ACT_MAX, nla, NULL, extack); + if (err < 0) + goto err_out; + + err = -EINVAL; + kind = tb[TCA_ACT_KIND]; + ops = tc_lookup_action(kind); + if (!ops) { /*some idjot trying to flush unknown action */ + NL_SET_ERR_MSG(extack, "Cannot flush unknown TC action"); + goto err_out; + } + + nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, + sizeof(*t), 0); + if (!nlh) { + NL_SET_ERR_MSG(extack, "Failed to create TC action flush notification"); + goto out_module_put; + } + t = nlmsg_data(nlh); + t->tca_family = AF_UNSPEC; + t->tca__pad1 = 0; + t->tca__pad2 = 0; + + nest = nla_nest_start(skb, TCA_ACT_TAB); + if (!nest) { + NL_SET_ERR_MSG(extack, "Failed to add new netlink message"); + goto out_module_put; + } + + err = ops->walk(net, skb, &dcb, RTM_DELACTION, ops, extack); + if (err <= 0) { + nla_nest_cancel(skb, nest); + goto out_module_put; + } + + nla_nest_end(skb, nest); + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + nlh->nlmsg_flags |= NLM_F_ROOT; + module_put(ops->owner); + err = rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); + if (err > 0) + return 0; + if (err < 0) + NL_SET_ERR_MSG(extack, "Failed to send TC action flush notification"); + + return err; + +out_module_put: + module_put(ops->owner); +err_out: + kfree_skb(skb); + return err; +} + +static int tcf_action_delete(struct net *net, struct tc_action *actions[]) +{ + int i; + + for (i = 0; i < TCA_ACT_MAX_PRIO && actions[i]; i++) { + struct tc_action *a = actions[i]; + const struct tc_action_ops *ops = a->ops; + /* Actions can be deleted concurrently so we must save their + * type and id to search again after reference is released. + */ + struct tcf_idrinfo *idrinfo = a->idrinfo; + u32 act_index = a->tcfa_index; + + actions[i] = NULL; + if (tcf_action_put(a)) { + /* last reference, action was deleted concurrently */ + module_put(ops->owner); + } else { + int ret; + + /* now do the delete */ + ret = tcf_idr_delete_index(idrinfo, act_index); + if (ret < 0) + return ret; + } + } + return 0; +} + +static int +tcf_del_notify(struct net *net, struct nlmsghdr *n, struct tc_action *actions[], + u32 portid, size_t attr_size, struct netlink_ext_ack *extack) +{ + int ret; + struct sk_buff *skb; + + skb = alloc_skb(attr_size <= NLMSG_GOODSIZE ? NLMSG_GOODSIZE : attr_size, + GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, RTM_DELACTION, + 0, 2) <= 0) { + NL_SET_ERR_MSG(extack, "Failed to fill netlink TC action attributes"); + kfree_skb(skb); + return -EINVAL; + } + + /* now do the delete */ + ret = tcf_action_delete(net, actions); + if (ret < 0) { + NL_SET_ERR_MSG(extack, "Failed to delete TC action"); + kfree_skb(skb); + return ret; + } + + ret = rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); + if (ret > 0) + return 0; + return ret; +} + +static int +tca_action_gd(struct net *net, struct nlattr *nla, struct nlmsghdr *n, + u32 portid, int event, struct netlink_ext_ack *extack) +{ + int i, ret; + struct nlattr *tb[TCA_ACT_MAX_PRIO + 1]; + struct tc_action *act; + size_t attr_size = 0; + struct tc_action *actions[TCA_ACT_MAX_PRIO] = {}; + + ret = nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL, extack); + if (ret < 0) + return ret; + + if (event == RTM_DELACTION && n->nlmsg_flags & NLM_F_ROOT) { + if (tb[1]) + return tca_action_flush(net, tb[1], n, portid, extack); + + NL_SET_ERR_MSG(extack, "Invalid netlink attributes while flushing TC action"); + return -EINVAL; + } + + for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) { + act = tcf_action_get_1(net, tb[i], n, portid, extack); + if (IS_ERR(act)) { + ret = PTR_ERR(act); + goto err; + } + attr_size += tcf_action_fill_size(act); + actions[i - 1] = act; + } + + attr_size = tcf_action_full_attrs_size(attr_size); + + if (event == RTM_GETACTION) + ret = tcf_get_notify(net, portid, n, actions, event, extack); + else { /* delete */ + ret = tcf_del_notify(net, n, actions, portid, attr_size, extack); + if (ret) + goto err; + return 0; + } +err: + tcf_action_put_many(actions); + return ret; +} + +static int +tcf_add_notify(struct net *net, struct nlmsghdr *n, struct tc_action *actions[], + u32 portid, size_t attr_size, struct netlink_ext_ack *extack) +{ + struct sk_buff *skb; + int err = 0; + + skb = alloc_skb(attr_size <= NLMSG_GOODSIZE ? NLMSG_GOODSIZE : attr_size, + GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, n->nlmsg_flags, + RTM_NEWACTION, 0, 0) <= 0) { + NL_SET_ERR_MSG(extack, "Failed to fill netlink attributes while adding TC action"); + kfree_skb(skb); + return -EINVAL; + } + + err = rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); + if (err > 0) + err = 0; + return err; +} + +static int tcf_action_add(struct net *net, struct nlattr *nla, + struct nlmsghdr *n, u32 portid, int ovr, + struct netlink_ext_ack *extack) +{ + size_t attr_size = 0; + int loop, ret; + struct tc_action *actions[TCA_ACT_MAX_PRIO] = {}; + + for (loop = 0; loop < 10; loop++) { + ret = tcf_action_init(net, NULL, nla, NULL, NULL, ovr, 0, + actions, &attr_size, true, extack); + if (ret != -EAGAIN) + break; + } + + if (ret < 0) + return ret; + ret = tcf_add_notify(net, n, actions, portid, attr_size, extack); + if (ovr) + tcf_action_put_many(actions); + + return ret; +} + +static u32 tcaa_root_flags_allowed = TCA_FLAG_LARGE_DUMP_ON; +static const struct nla_policy tcaa_policy[TCA_ROOT_MAX + 1] = { + [TCA_ROOT_FLAGS] = { .type = NLA_BITFIELD32, + .validation_data = &tcaa_root_flags_allowed }, + [TCA_ROOT_TIME_DELTA] = { .type = NLA_U32 }, +}; + +static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_ROOT_MAX + 1]; + u32 portid = skb ? NETLINK_CB(skb).portid : 0; + int ret = 0, ovr = 0; + + if ((n->nlmsg_type != RTM_GETACTION) && + !netlink_capable(skb, CAP_NET_ADMIN)) + return -EPERM; + + ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ROOT_MAX, NULL, + extack); + if (ret < 0) + return ret; + + if (tca[TCA_ACT_TAB] == NULL) { + NL_SET_ERR_MSG(extack, "Netlink action attributes missing"); + return -EINVAL; + } + + /* n->nlmsg_flags & NLM_F_CREATE */ + switch (n->nlmsg_type) { + case RTM_NEWACTION: + /* we are going to assume all other flags + * imply create only if it doesn't exist + * Note that CREATE | EXCL implies that + * but since we want avoid ambiguity (eg when flags + * is zero) then just set this + */ + if (n->nlmsg_flags & NLM_F_REPLACE) + ovr = 1; + ret = tcf_action_add(net, tca[TCA_ACT_TAB], n, portid, ovr, + extack); + break; + case RTM_DELACTION: + ret = tca_action_gd(net, tca[TCA_ACT_TAB], n, + portid, RTM_DELACTION, extack); + break; + case RTM_GETACTION: + ret = tca_action_gd(net, tca[TCA_ACT_TAB], n, + portid, RTM_GETACTION, extack); + break; + default: + BUG(); + } + + return ret; +} + +static struct nlattr *find_dump_kind(struct nlattr **nla) +{ + struct nlattr *tb1, *tb2[TCA_ACT_MAX + 1]; + struct nlattr *tb[TCA_ACT_MAX_PRIO + 1]; + struct nlattr *kind; + + tb1 = nla[TCA_ACT_TAB]; + if (tb1 == NULL) + return NULL; + + if (nla_parse(tb, TCA_ACT_MAX_PRIO, nla_data(tb1), + NLMSG_ALIGN(nla_len(tb1)), NULL, NULL) < 0) + return NULL; + + if (tb[1] == NULL) + return NULL; + if (nla_parse_nested(tb2, TCA_ACT_MAX, tb[1], NULL, NULL) < 0) + return NULL; + kind = tb2[TCA_ACT_KIND]; + + return kind; +} + +static int tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + struct nlattr *nest; + struct tc_action_ops *a_o; + int ret = 0; + struct tcamsg *t = (struct tcamsg *) nlmsg_data(cb->nlh); + struct nlattr *tb[TCA_ROOT_MAX + 1]; + struct nlattr *count_attr = NULL; + unsigned long jiffy_since = 0; + struct nlattr *kind = NULL; + struct nla_bitfield32 bf; + u32 msecs_since = 0; + u32 act_count = 0; + + ret = nlmsg_parse(cb->nlh, sizeof(struct tcamsg), tb, TCA_ROOT_MAX, + tcaa_policy, NULL); + if (ret < 0) + return ret; + + kind = find_dump_kind(tb); + if (kind == NULL) { + pr_info("tc_dump_action: action bad kind\n"); + return 0; + } + + a_o = tc_lookup_action(kind); + if (a_o == NULL) + return 0; + + cb->args[2] = 0; + if (tb[TCA_ROOT_FLAGS]) { + bf = nla_get_bitfield32(tb[TCA_ROOT_FLAGS]); + cb->args[2] = bf.value; + } + + if (tb[TCA_ROOT_TIME_DELTA]) { + msecs_since = nla_get_u32(tb[TCA_ROOT_TIME_DELTA]); + } + + nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, + cb->nlh->nlmsg_type, sizeof(*t), 0); + if (!nlh) + goto out_module_put; + + if (msecs_since) + jiffy_since = jiffies - msecs_to_jiffies(msecs_since); + + t = nlmsg_data(nlh); + t->tca_family = AF_UNSPEC; + t->tca__pad1 = 0; + t->tca__pad2 = 0; + cb->args[3] = jiffy_since; + count_attr = nla_reserve(skb, TCA_ROOT_COUNT, sizeof(u32)); + if (!count_attr) + goto out_module_put; + + nest = nla_nest_start(skb, TCA_ACT_TAB); + if (nest == NULL) + goto out_module_put; + + ret = a_o->walk(net, skb, cb, RTM_GETACTION, a_o, NULL); + if (ret < 0) + goto out_module_put; + + if (ret > 0) { + nla_nest_end(skb, nest); + ret = skb->len; + act_count = cb->args[1]; + memcpy(nla_data(count_attr), &act_count, sizeof(u32)); + cb->args[1] = 0; + } else + nlmsg_trim(skb, b); + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + if (NETLINK_CB(cb->skb).portid && ret) + nlh->nlmsg_flags |= NLM_F_MULTI; + module_put(a_o->owner); + return skb->len; + +out_module_put: + module_put(a_o->owner); + nlmsg_trim(skb, b); + return skb->len; +} + +struct tcf_action_net { + struct rhashtable egdev_ht; +}; + +static unsigned int tcf_action_net_id; + +struct tcf_action_egdev_cb { + struct list_head list; + tc_setup_cb_t *cb; + void *cb_priv; +}; + +struct tcf_action_egdev { + struct rhash_head ht_node; + const struct net_device *dev; + unsigned int refcnt; + struct list_head cb_list; +}; + +static const struct rhashtable_params tcf_action_egdev_ht_params = { + .key_offset = offsetof(struct tcf_action_egdev, dev), + .head_offset = offsetof(struct tcf_action_egdev, ht_node), + .key_len = sizeof(const struct net_device *), +}; + +static struct tcf_action_egdev * +tcf_action_egdev_lookup(const struct net_device *dev) +{ + struct net *net = dev_net(dev); + struct tcf_action_net *tan = net_generic(net, tcf_action_net_id); + + return rhashtable_lookup_fast(&tan->egdev_ht, &dev, + tcf_action_egdev_ht_params); +} + +static struct tcf_action_egdev * +tcf_action_egdev_get(const struct net_device *dev) +{ + struct tcf_action_egdev *egdev; + struct tcf_action_net *tan; + + egdev = tcf_action_egdev_lookup(dev); + if (egdev) + goto inc_ref; + + egdev = kzalloc(sizeof(*egdev), GFP_KERNEL); + if (!egdev) + return NULL; + INIT_LIST_HEAD(&egdev->cb_list); + egdev->dev = dev; + tan = net_generic(dev_net(dev), tcf_action_net_id); + rhashtable_insert_fast(&tan->egdev_ht, &egdev->ht_node, + tcf_action_egdev_ht_params); + +inc_ref: + egdev->refcnt++; + return egdev; +} + +static void tcf_action_egdev_put(struct tcf_action_egdev *egdev) +{ + struct tcf_action_net *tan; + + if (--egdev->refcnt) + return; + tan = net_generic(dev_net(egdev->dev), tcf_action_net_id); + rhashtable_remove_fast(&tan->egdev_ht, &egdev->ht_node, + tcf_action_egdev_ht_params); + kfree(egdev); +} + +static struct tcf_action_egdev_cb * +tcf_action_egdev_cb_lookup(struct tcf_action_egdev *egdev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev_cb *egdev_cb; + + list_for_each_entry(egdev_cb, &egdev->cb_list, list) + if (egdev_cb->cb == cb && egdev_cb->cb_priv == cb_priv) + return egdev_cb; + return NULL; +} + +static int tcf_action_egdev_cb_call(struct tcf_action_egdev *egdev, + enum tc_setup_type type, + void *type_data, bool err_stop) +{ + struct tcf_action_egdev_cb *egdev_cb; + int ok_count = 0; + int err; + + list_for_each_entry(egdev_cb, &egdev->cb_list, list) { + err = egdev_cb->cb(type, type_data, egdev_cb->cb_priv); + if (err) { + if (err_stop) + return err; + } else { + ok_count++; + } + } + return ok_count; +} + +static int tcf_action_egdev_cb_add(struct tcf_action_egdev *egdev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev_cb *egdev_cb; + + egdev_cb = tcf_action_egdev_cb_lookup(egdev, cb, cb_priv); + if (WARN_ON(egdev_cb)) + return -EEXIST; + egdev_cb = kzalloc(sizeof(*egdev_cb), GFP_KERNEL); + if (!egdev_cb) + return -ENOMEM; + egdev_cb->cb = cb; + egdev_cb->cb_priv = cb_priv; + list_add(&egdev_cb->list, &egdev->cb_list); + return 0; +} + +static void tcf_action_egdev_cb_del(struct tcf_action_egdev *egdev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev_cb *egdev_cb; + + egdev_cb = tcf_action_egdev_cb_lookup(egdev, cb, cb_priv); + if (WARN_ON(!egdev_cb)) + return; + list_del(&egdev_cb->list); + kfree(egdev_cb); +} + +static int __tc_setup_cb_egdev_register(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev *egdev = tcf_action_egdev_get(dev); + int err; + + if (!egdev) + return -ENOMEM; + err = tcf_action_egdev_cb_add(egdev, cb, cb_priv); + if (err) + goto err_cb_add; + return 0; + +err_cb_add: + tcf_action_egdev_put(egdev); + return err; +} +int tc_setup_cb_egdev_register(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + int err; + + rtnl_lock(); + err = __tc_setup_cb_egdev_register(dev, cb, cb_priv); + rtnl_unlock(); + return err; +} +EXPORT_SYMBOL_GPL(tc_setup_cb_egdev_register); + +static void __tc_setup_cb_egdev_unregister(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev *egdev = tcf_action_egdev_lookup(dev); + + if (WARN_ON(!egdev)) + return; + tcf_action_egdev_cb_del(egdev, cb, cb_priv); + tcf_action_egdev_put(egdev); +} +void tc_setup_cb_egdev_unregister(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + rtnl_lock(); + __tc_setup_cb_egdev_unregister(dev, cb, cb_priv); + rtnl_unlock(); +} +EXPORT_SYMBOL_GPL(tc_setup_cb_egdev_unregister); + +int tc_setup_cb_egdev_call(const struct net_device *dev, + enum tc_setup_type type, void *type_data, + bool err_stop) +{ + struct tcf_action_egdev *egdev = tcf_action_egdev_lookup(dev); + + if (!egdev) + return 0; + return tcf_action_egdev_cb_call(egdev, type, type_data, err_stop); +} +EXPORT_SYMBOL_GPL(tc_setup_cb_egdev_call); + +static __net_init int tcf_action_net_init(struct net *net) +{ + struct tcf_action_net *tan = net_generic(net, tcf_action_net_id); + + return rhashtable_init(&tan->egdev_ht, &tcf_action_egdev_ht_params); +} + +static void __net_exit tcf_action_net_exit(struct net *net) +{ + struct tcf_action_net *tan = net_generic(net, tcf_action_net_id); + + rhashtable_destroy(&tan->egdev_ht); +} + +static struct pernet_operations tcf_action_net_ops = { + .init = tcf_action_net_init, + .exit = tcf_action_net_exit, + .id = &tcf_action_net_id, + .size = sizeof(struct tcf_action_net), +}; + +static int __init tc_action_init(void) +{ + int err; + + err = register_pernet_subsys(&tcf_action_net_ops); + if (err) + return err; + + rtnl_register(PF_UNSPEC, RTM_NEWACTION, tc_ctl_action, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELACTION, tc_ctl_action, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETACTION, tc_ctl_action, tc_dump_action, + 0); + + return 0; +} + +subsys_initcall(tc_action_init); diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c new file mode 100644 index 000000000..800846d77 --- /dev/null +++ b/net/sched/act_bpf.c @@ -0,0 +1,446 @@ +/* + * Copyright (c) 2015 Jiri Pirko <jiri@resnulli.us> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/filter.h> +#include <linux/bpf.h> + +#include <net/netlink.h> +#include <net/pkt_sched.h> + +#include <linux/tc_act/tc_bpf.h> +#include <net/tc_act/tc_bpf.h> + +#define ACT_BPF_NAME_LEN 256 + +struct tcf_bpf_cfg { + struct bpf_prog *filter; + struct sock_filter *bpf_ops; + const char *bpf_name; + u16 bpf_num_ops; + bool is_ebpf; +}; + +static unsigned int bpf_net_id; +static struct tc_action_ops act_bpf_ops; + +static int tcf_bpf_act(struct sk_buff *skb, const struct tc_action *act, + struct tcf_result *res) +{ + bool at_ingress = skb_at_tc_ingress(skb); + struct tcf_bpf *prog = to_bpf(act); + struct bpf_prog *filter; + int action, filter_res; + + tcf_lastuse_update(&prog->tcf_tm); + bstats_cpu_update(this_cpu_ptr(prog->common.cpu_bstats), skb); + + rcu_read_lock(); + filter = rcu_dereference(prog->filter); + if (at_ingress) { + __skb_push(skb, skb->mac_len); + bpf_compute_data_pointers(skb); + filter_res = BPF_PROG_RUN(filter, skb); + __skb_pull(skb, skb->mac_len); + } else { + bpf_compute_data_pointers(skb); + filter_res = BPF_PROG_RUN(filter, skb); + } + rcu_read_unlock(); + + /* A BPF program may overwrite the default action opcode. + * Similarly as in cls_bpf, if filter_res == -1 we use the + * default action specified from tc. + * + * In case a different well-known TC_ACT opcode has been + * returned, it will overwrite the default one. + * + * For everything else that is unkown, TC_ACT_UNSPEC is + * returned. + */ + switch (filter_res) { + case TC_ACT_PIPE: + case TC_ACT_RECLASSIFY: + case TC_ACT_OK: + case TC_ACT_REDIRECT: + action = filter_res; + break; + case TC_ACT_SHOT: + action = filter_res; + qstats_drop_inc(this_cpu_ptr(prog->common.cpu_qstats)); + break; + case TC_ACT_UNSPEC: + action = prog->tcf_action; + break; + default: + action = TC_ACT_UNSPEC; + break; + } + + return action; +} + +static bool tcf_bpf_is_ebpf(const struct tcf_bpf *prog) +{ + return !prog->bpf_ops; +} + +static int tcf_bpf_dump_bpf_info(const struct tcf_bpf *prog, + struct sk_buff *skb) +{ + struct nlattr *nla; + + if (nla_put_u16(skb, TCA_ACT_BPF_OPS_LEN, prog->bpf_num_ops)) + return -EMSGSIZE; + + nla = nla_reserve(skb, TCA_ACT_BPF_OPS, prog->bpf_num_ops * + sizeof(struct sock_filter)); + if (nla == NULL) + return -EMSGSIZE; + + memcpy(nla_data(nla), prog->bpf_ops, nla_len(nla)); + + return 0; +} + +static int tcf_bpf_dump_ebpf_info(const struct tcf_bpf *prog, + struct sk_buff *skb) +{ + struct nlattr *nla; + + if (prog->bpf_name && + nla_put_string(skb, TCA_ACT_BPF_NAME, prog->bpf_name)) + return -EMSGSIZE; + + if (nla_put_u32(skb, TCA_ACT_BPF_ID, prog->filter->aux->id)) + return -EMSGSIZE; + + nla = nla_reserve(skb, TCA_ACT_BPF_TAG, sizeof(prog->filter->tag)); + if (nla == NULL) + return -EMSGSIZE; + + memcpy(nla_data(nla), prog->filter->tag, nla_len(nla)); + + return 0; +} + +static int tcf_bpf_dump(struct sk_buff *skb, struct tc_action *act, + int bind, int ref) +{ + unsigned char *tp = skb_tail_pointer(skb); + struct tcf_bpf *prog = to_bpf(act); + struct tc_act_bpf opt = { + .index = prog->tcf_index, + .refcnt = refcount_read(&prog->tcf_refcnt) - ref, + .bindcnt = atomic_read(&prog->tcf_bindcnt) - bind, + }; + struct tcf_t tm; + int ret; + + spin_lock_bh(&prog->tcf_lock); + opt.action = prog->tcf_action; + if (nla_put(skb, TCA_ACT_BPF_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + if (tcf_bpf_is_ebpf(prog)) + ret = tcf_bpf_dump_ebpf_info(prog, skb); + else + ret = tcf_bpf_dump_bpf_info(prog, skb); + if (ret) + goto nla_put_failure; + + tcf_tm_dump(&tm, &prog->tcf_tm); + if (nla_put_64bit(skb, TCA_ACT_BPF_TM, sizeof(tm), &tm, + TCA_ACT_BPF_PAD)) + goto nla_put_failure; + + spin_unlock_bh(&prog->tcf_lock); + return skb->len; + +nla_put_failure: + spin_unlock_bh(&prog->tcf_lock); + nlmsg_trim(skb, tp); + return -1; +} + +static const struct nla_policy act_bpf_policy[TCA_ACT_BPF_MAX + 1] = { + [TCA_ACT_BPF_PARMS] = { .len = sizeof(struct tc_act_bpf) }, + [TCA_ACT_BPF_FD] = { .type = NLA_U32 }, + [TCA_ACT_BPF_NAME] = { .type = NLA_NUL_STRING, + .len = ACT_BPF_NAME_LEN }, + [TCA_ACT_BPF_OPS_LEN] = { .type = NLA_U16 }, + [TCA_ACT_BPF_OPS] = { .type = NLA_BINARY, + .len = sizeof(struct sock_filter) * BPF_MAXINSNS }, +}; + +static int tcf_bpf_init_from_ops(struct nlattr **tb, struct tcf_bpf_cfg *cfg) +{ + struct sock_filter *bpf_ops; + struct sock_fprog_kern fprog_tmp; + struct bpf_prog *fp; + u16 bpf_size, bpf_num_ops; + int ret; + + bpf_num_ops = nla_get_u16(tb[TCA_ACT_BPF_OPS_LEN]); + if (bpf_num_ops > BPF_MAXINSNS || bpf_num_ops == 0) + return -EINVAL; + + bpf_size = bpf_num_ops * sizeof(*bpf_ops); + if (bpf_size != nla_len(tb[TCA_ACT_BPF_OPS])) + return -EINVAL; + + bpf_ops = kmemdup(nla_data(tb[TCA_ACT_BPF_OPS]), bpf_size, GFP_KERNEL); + if (bpf_ops == NULL) + return -ENOMEM; + + fprog_tmp.len = bpf_num_ops; + fprog_tmp.filter = bpf_ops; + + ret = bpf_prog_create(&fp, &fprog_tmp); + if (ret < 0) { + kfree(bpf_ops); + return ret; + } + + cfg->bpf_ops = bpf_ops; + cfg->bpf_num_ops = bpf_num_ops; + cfg->filter = fp; + cfg->is_ebpf = false; + + return 0; +} + +static int tcf_bpf_init_from_efd(struct nlattr **tb, struct tcf_bpf_cfg *cfg) +{ + struct bpf_prog *fp; + char *name = NULL; + u32 bpf_fd; + + bpf_fd = nla_get_u32(tb[TCA_ACT_BPF_FD]); + + fp = bpf_prog_get_type(bpf_fd, BPF_PROG_TYPE_SCHED_ACT); + if (IS_ERR(fp)) + return PTR_ERR(fp); + + if (tb[TCA_ACT_BPF_NAME]) { + name = nla_memdup(tb[TCA_ACT_BPF_NAME], GFP_KERNEL); + if (!name) { + bpf_prog_put(fp); + return -ENOMEM; + } + } + + cfg->bpf_name = name; + cfg->filter = fp; + cfg->is_ebpf = true; + + return 0; +} + +static void tcf_bpf_cfg_cleanup(const struct tcf_bpf_cfg *cfg) +{ + struct bpf_prog *filter = cfg->filter; + + if (filter) { + if (cfg->is_ebpf) + bpf_prog_put(filter); + else + bpf_prog_destroy(filter); + } + + kfree(cfg->bpf_ops); + kfree(cfg->bpf_name); +} + +static void tcf_bpf_prog_fill_cfg(const struct tcf_bpf *prog, + struct tcf_bpf_cfg *cfg) +{ + cfg->is_ebpf = tcf_bpf_is_ebpf(prog); + /* updates to prog->filter are prevented, since it's called either + * with tcf lock or during final cleanup in rcu callback + */ + cfg->filter = rcu_dereference_protected(prog->filter, 1); + + cfg->bpf_ops = prog->bpf_ops; + cfg->bpf_name = prog->bpf_name; +} + +static int tcf_bpf_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **act, + int replace, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, bpf_net_id); + struct nlattr *tb[TCA_ACT_BPF_MAX + 1]; + struct tcf_bpf_cfg cfg, old; + struct tc_act_bpf *parm; + struct tcf_bpf *prog; + bool is_bpf, is_ebpf; + int ret, res = 0; + u32 index; + + if (!nla) + return -EINVAL; + + ret = nla_parse_nested(tb, TCA_ACT_BPF_MAX, nla, act_bpf_policy, NULL); + if (ret < 0) + return ret; + + if (!tb[TCA_ACT_BPF_PARMS]) + return -EINVAL; + + parm = nla_data(tb[TCA_ACT_BPF_PARMS]); + index = parm->index; + ret = tcf_idr_check_alloc(tn, &index, act, bind); + if (!ret) { + ret = tcf_idr_create(tn, index, est, act, + &act_bpf_ops, bind, true); + if (ret < 0) { + tcf_idr_cleanup(tn, index); + return ret; + } + + res = ACT_P_CREATED; + } else if (ret > 0) { + /* Don't override defaults. */ + if (bind) + return 0; + + if (!replace) { + tcf_idr_release(*act, bind); + return -EEXIST; + } + } else { + return ret; + } + + is_bpf = tb[TCA_ACT_BPF_OPS_LEN] && tb[TCA_ACT_BPF_OPS]; + is_ebpf = tb[TCA_ACT_BPF_FD]; + + if ((!is_bpf && !is_ebpf) || (is_bpf && is_ebpf)) { + ret = -EINVAL; + goto out; + } + + memset(&cfg, 0, sizeof(cfg)); + + ret = is_bpf ? tcf_bpf_init_from_ops(tb, &cfg) : + tcf_bpf_init_from_efd(tb, &cfg); + if (ret < 0) + goto out; + + prog = to_bpf(*act); + + spin_lock_bh(&prog->tcf_lock); + if (res != ACT_P_CREATED) + tcf_bpf_prog_fill_cfg(prog, &old); + + prog->bpf_ops = cfg.bpf_ops; + prog->bpf_name = cfg.bpf_name; + + if (cfg.bpf_num_ops) + prog->bpf_num_ops = cfg.bpf_num_ops; + + prog->tcf_action = parm->action; + rcu_assign_pointer(prog->filter, cfg.filter); + spin_unlock_bh(&prog->tcf_lock); + + if (res == ACT_P_CREATED) { + tcf_idr_insert(tn, *act); + } else { + /* make sure the program being replaced is no longer executing */ + synchronize_rcu(); + tcf_bpf_cfg_cleanup(&old); + } + + return res; +out: + tcf_idr_release(*act, bind); + + return ret; +} + +static void tcf_bpf_cleanup(struct tc_action *act) +{ + struct tcf_bpf_cfg tmp; + + tcf_bpf_prog_fill_cfg(to_bpf(act), &tmp); + tcf_bpf_cfg_cleanup(&tmp); +} + +static int tcf_bpf_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, bpf_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_bpf_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, bpf_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_bpf_ops __read_mostly = { + .kind = "bpf", + .type = TCA_ACT_BPF, + .owner = THIS_MODULE, + .act = tcf_bpf_act, + .dump = tcf_bpf_dump, + .cleanup = tcf_bpf_cleanup, + .init = tcf_bpf_init, + .walk = tcf_bpf_walker, + .lookup = tcf_bpf_search, + .size = sizeof(struct tcf_bpf), +}; + +static __net_init int bpf_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, bpf_net_id); + + return tc_action_net_init(net, tn, &act_bpf_ops); +} + +static void __net_exit bpf_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, bpf_net_id); +} + +static struct pernet_operations bpf_net_ops = { + .init = bpf_init_net, + .exit_batch = bpf_exit_net, + .id = &bpf_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init bpf_init_module(void) +{ + return tcf_register_action(&act_bpf_ops, &bpf_net_ops); +} + +static void __exit bpf_cleanup_module(void) +{ + tcf_unregister_action(&act_bpf_ops, &bpf_net_ops); +} + +module_init(bpf_init_module); +module_exit(bpf_cleanup_module); + +MODULE_AUTHOR("Jiri Pirko <jiri@resnulli.us>"); +MODULE_DESCRIPTION("TC BPF based action"); +MODULE_LICENSE("GPL v2"); diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c new file mode 100644 index 000000000..6485f421c --- /dev/null +++ b/net/sched/act_connmark.c @@ -0,0 +1,250 @@ +/* + * net/sched/act_connmark.c netfilter connmark retriever action + * skb mark is over-written + * + * Copyright (c) 2011 Felix Fietkau <nbd@openwrt.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. +*/ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/pkt_cls.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/act_api.h> +#include <uapi/linux/tc_act/tc_connmark.h> +#include <net/tc_act/tc_connmark.h> + +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/nf_conntrack_zones.h> + +static unsigned int connmark_net_id; +static struct tc_action_ops act_connmark_ops; + +static int tcf_connmark_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + const struct nf_conntrack_tuple_hash *thash; + struct nf_conntrack_tuple tuple; + enum ip_conntrack_info ctinfo; + struct tcf_connmark_info *ca = to_connmark(a); + struct nf_conntrack_zone zone; + struct nf_conn *c; + int proto; + + spin_lock(&ca->tcf_lock); + tcf_lastuse_update(&ca->tcf_tm); + bstats_update(&ca->tcf_bstats, skb); + + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + if (skb->len < sizeof(struct iphdr)) + goto out; + + proto = NFPROTO_IPV4; + break; + case htons(ETH_P_IPV6): + if (skb->len < sizeof(struct ipv6hdr)) + goto out; + + proto = NFPROTO_IPV6; + break; + default: + goto out; + } + + c = nf_ct_get(skb, &ctinfo); + if (c) { + skb->mark = c->mark; + /* using overlimits stats to count how many packets marked */ + ca->tcf_qstats.overlimits++; + goto out; + } + + if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), + proto, ca->net, &tuple)) + goto out; + + zone.id = ca->zone; + zone.dir = NF_CT_DEFAULT_ZONE_DIR; + + thash = nf_conntrack_find_get(ca->net, &zone, &tuple); + if (!thash) + goto out; + + c = nf_ct_tuplehash_to_ctrack(thash); + /* using overlimits stats to count how many packets marked */ + ca->tcf_qstats.overlimits++; + skb->mark = c->mark; + nf_ct_put(c); + +out: + spin_unlock(&ca->tcf_lock); + return ca->tcf_action; +} + +static const struct nla_policy connmark_policy[TCA_CONNMARK_MAX + 1] = { + [TCA_CONNMARK_PARMS] = { .len = sizeof(struct tc_connmark) }, +}; + +static int tcf_connmark_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, connmark_net_id); + struct nlattr *tb[TCA_CONNMARK_MAX + 1]; + struct tcf_connmark_info *ci; + struct tc_connmark *parm; + int ret = 0; + u32 index; + + if (!nla) + return -EINVAL; + + ret = nla_parse_nested(tb, TCA_CONNMARK_MAX, nla, connmark_policy, + NULL); + if (ret < 0) + return ret; + + if (!tb[TCA_CONNMARK_PARMS]) + return -EINVAL; + + parm = nla_data(tb[TCA_CONNMARK_PARMS]); + index = parm->index; + ret = tcf_idr_check_alloc(tn, &index, a, bind); + if (!ret) { + ret = tcf_idr_create(tn, index, est, a, + &act_connmark_ops, bind, false); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + + ci = to_connmark(*a); + ci->tcf_action = parm->action; + ci->net = net; + ci->zone = parm->zone; + + tcf_idr_insert(tn, *a); + ret = ACT_P_CREATED; + } else if (ret > 0) { + ci = to_connmark(*a); + if (bind) + return 0; + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + /* replacing action and zone */ + ci->tcf_action = parm->action; + ci->zone = parm->zone; + ret = 0; + } + + return ret; +} + +static inline int tcf_connmark_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_connmark_info *ci = to_connmark(a); + + struct tc_connmark opt = { + .index = ci->tcf_index, + .refcnt = refcount_read(&ci->tcf_refcnt) - ref, + .bindcnt = atomic_read(&ci->tcf_bindcnt) - bind, + .action = ci->tcf_action, + .zone = ci->zone, + }; + struct tcf_t t; + + if (nla_put(skb, TCA_CONNMARK_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + tcf_tm_dump(&t, &ci->tcf_tm); + if (nla_put_64bit(skb, TCA_CONNMARK_TM, sizeof(t), &t, + TCA_CONNMARK_PAD)) + goto nla_put_failure; + + return skb->len; +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_connmark_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, connmark_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_connmark_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, connmark_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_connmark_ops = { + .kind = "connmark", + .type = TCA_ACT_CONNMARK, + .owner = THIS_MODULE, + .act = tcf_connmark_act, + .dump = tcf_connmark_dump, + .init = tcf_connmark_init, + .walk = tcf_connmark_walker, + .lookup = tcf_connmark_search, + .size = sizeof(struct tcf_connmark_info), +}; + +static __net_init int connmark_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, connmark_net_id); + + return tc_action_net_init(net, tn, &act_connmark_ops); +} + +static void __net_exit connmark_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, connmark_net_id); +} + +static struct pernet_operations connmark_net_ops = { + .init = connmark_init_net, + .exit_batch = connmark_exit_net, + .id = &connmark_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init connmark_init_module(void) +{ + return tcf_register_action(&act_connmark_ops, &connmark_net_ops); +} + +static void __exit connmark_cleanup_module(void) +{ + tcf_unregister_action(&act_connmark_ops, &connmark_net_ops); +} + +module_init(connmark_init_module); +module_exit(connmark_cleanup_module); +MODULE_AUTHOR("Felix Fietkau <nbd@openwrt.org>"); +MODULE_DESCRIPTION("Connection tracking mark restoring"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c new file mode 100644 index 000000000..24ad4ceac --- /dev/null +++ b/net/sched/act_csum.c @@ -0,0 +1,737 @@ +/* + * Checksum updating actions + * + * Copyright (c) 2010 Gregoire Baron <baronchon@n7mm.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. + * + */ + +#include <linux/types.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/spinlock.h> + +#include <linux/netlink.h> +#include <net/netlink.h> +#include <linux/rtnetlink.h> + +#include <linux/skbuff.h> + +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/icmp.h> +#include <linux/icmpv6.h> +#include <linux/igmp.h> +#include <net/tcp.h> +#include <net/udp.h> +#include <net/ip6_checksum.h> +#include <net/sctp/checksum.h> + +#include <net/act_api.h> + +#include <linux/tc_act/tc_csum.h> +#include <net/tc_act/tc_csum.h> + +static const struct nla_policy csum_policy[TCA_CSUM_MAX + 1] = { + [TCA_CSUM_PARMS] = { .len = sizeof(struct tc_csum), }, +}; + +static unsigned int csum_net_id; +static struct tc_action_ops act_csum_ops; + +static int tcf_csum_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, int ovr, + int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, csum_net_id); + struct tcf_csum_params *params_new; + struct nlattr *tb[TCA_CSUM_MAX + 1]; + struct tc_csum *parm; + struct tcf_csum *p; + int ret = 0, err; + u32 index; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_CSUM_MAX, nla, csum_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_CSUM_PARMS] == NULL) + return -EINVAL; + parm = nla_data(tb[TCA_CSUM_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (!err) { + ret = tcf_idr_create(tn, index, est, a, + &act_csum_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else if (err > 0) { + if (bind)/* dont override defaults */ + return 0; + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + } else { + return err; + } + + p = to_tcf_csum(*a); + + params_new = kzalloc(sizeof(*params_new), GFP_KERNEL); + if (unlikely(!params_new)) { + tcf_idr_release(*a, bind); + return -ENOMEM; + } + params_new->update_flags = parm->update_flags; + + spin_lock_bh(&p->tcf_lock); + p->tcf_action = parm->action; + rcu_swap_protected(p->params, params_new, + lockdep_is_held(&p->tcf_lock)); + spin_unlock_bh(&p->tcf_lock); + + if (params_new) + kfree_rcu(params_new, rcu); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + + return ret; +} + +/** + * tcf_csum_skb_nextlayer - Get next layer pointer + * @skb: sk_buff to use + * @ihl: previous summed headers length + * @ipl: complete packet length + * @jhl: next header length + * + * Check the expected next layer availability in the specified sk_buff. + * Return the next layer pointer if pass, NULL otherwise. + */ +static void *tcf_csum_skb_nextlayer(struct sk_buff *skb, + unsigned int ihl, unsigned int ipl, + unsigned int jhl) +{ + int ntkoff = skb_network_offset(skb); + int hl = ihl + jhl; + + if (!pskb_may_pull(skb, ipl + ntkoff) || (ipl < hl) || + skb_try_make_writable(skb, hl + ntkoff)) + return NULL; + else + return (void *)(skb_network_header(skb) + ihl); +} + +static int tcf_csum_ipv4_icmp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl) +{ + struct icmphdr *icmph; + + icmph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*icmph)); + if (icmph == NULL) + return 0; + + icmph->checksum = 0; + skb->csum = csum_partial(icmph, ipl - ihl, 0); + icmph->checksum = csum_fold(skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv4_igmp(struct sk_buff *skb, + unsigned int ihl, unsigned int ipl) +{ + struct igmphdr *igmph; + + igmph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*igmph)); + if (igmph == NULL) + return 0; + + igmph->csum = 0; + skb->csum = csum_partial(igmph, ipl - ihl, 0); + igmph->csum = csum_fold(skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv6_icmp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl) +{ + struct icmp6hdr *icmp6h; + const struct ipv6hdr *ip6h; + + icmp6h = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*icmp6h)); + if (icmp6h == NULL) + return 0; + + ip6h = ipv6_hdr(skb); + icmp6h->icmp6_cksum = 0; + skb->csum = csum_partial(icmp6h, ipl - ihl, 0); + icmp6h->icmp6_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, + ipl - ihl, IPPROTO_ICMPV6, + skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv4_tcp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl) +{ + struct tcphdr *tcph; + const struct iphdr *iph; + + if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4) + return 1; + + tcph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*tcph)); + if (tcph == NULL) + return 0; + + iph = ip_hdr(skb); + tcph->check = 0; + skb->csum = csum_partial(tcph, ipl - ihl, 0); + tcph->check = tcp_v4_check(ipl - ihl, + iph->saddr, iph->daddr, skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv6_tcp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl) +{ + struct tcphdr *tcph; + const struct ipv6hdr *ip6h; + + if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) + return 1; + + tcph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*tcph)); + if (tcph == NULL) + return 0; + + ip6h = ipv6_hdr(skb); + tcph->check = 0; + skb->csum = csum_partial(tcph, ipl - ihl, 0); + tcph->check = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, + ipl - ihl, IPPROTO_TCP, + skb->csum); + + skb->ip_summed = CHECKSUM_NONE; + + return 1; +} + +static int tcf_csum_ipv4_udp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl, int udplite) +{ + struct udphdr *udph; + const struct iphdr *iph; + u16 ul; + + if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP) + return 1; + + /* + * Support both UDP and UDPLITE checksum algorithms, Don't use + * udph->len to get the real length without any protocol check, + * UDPLITE uses udph->len for another thing, + * Use iph->tot_len, or just ipl. + */ + + udph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*udph)); + if (udph == NULL) + return 0; + + iph = ip_hdr(skb); + ul = ntohs(udph->len); + + if (udplite || udph->check) { + + udph->check = 0; + + if (udplite) { + if (ul == 0) + skb->csum = csum_partial(udph, ipl - ihl, 0); + else if ((ul >= sizeof(*udph)) && (ul <= ipl - ihl)) + skb->csum = csum_partial(udph, ul, 0); + else + goto ignore_obscure_skb; + } else { + if (ul != ipl - ihl) + goto ignore_obscure_skb; + + skb->csum = csum_partial(udph, ul, 0); + } + + udph->check = csum_tcpudp_magic(iph->saddr, iph->daddr, + ul, iph->protocol, + skb->csum); + + if (!udph->check) + udph->check = CSUM_MANGLED_0; + } + + skb->ip_summed = CHECKSUM_NONE; + +ignore_obscure_skb: + return 1; +} + +static int tcf_csum_ipv6_udp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl, int udplite) +{ + struct udphdr *udph; + const struct ipv6hdr *ip6h; + u16 ul; + + if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP) + return 1; + + /* + * Support both UDP and UDPLITE checksum algorithms, Don't use + * udph->len to get the real length without any protocol check, + * UDPLITE uses udph->len for another thing, + * Use ip6h->payload_len + sizeof(*ip6h) ... , or just ipl. + */ + + udph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*udph)); + if (udph == NULL) + return 0; + + ip6h = ipv6_hdr(skb); + ul = ntohs(udph->len); + + udph->check = 0; + + if (udplite) { + if (ul == 0) + skb->csum = csum_partial(udph, ipl - ihl, 0); + + else if ((ul >= sizeof(*udph)) && (ul <= ipl - ihl)) + skb->csum = csum_partial(udph, ul, 0); + + else + goto ignore_obscure_skb; + } else { + if (ul != ipl - ihl) + goto ignore_obscure_skb; + + skb->csum = csum_partial(udph, ul, 0); + } + + udph->check = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, ul, + udplite ? IPPROTO_UDPLITE : IPPROTO_UDP, + skb->csum); + + if (!udph->check) + udph->check = CSUM_MANGLED_0; + + skb->ip_summed = CHECKSUM_NONE; + +ignore_obscure_skb: + return 1; +} + +static int tcf_csum_sctp(struct sk_buff *skb, unsigned int ihl, + unsigned int ipl) +{ + struct sctphdr *sctph; + + if (skb_is_gso(skb) && skb_is_gso_sctp(skb)) + return 1; + + sctph = tcf_csum_skb_nextlayer(skb, ihl, ipl, sizeof(*sctph)); + if (!sctph) + return 0; + + sctph->checksum = sctp_compute_cksum(skb, + skb_network_offset(skb) + ihl); + skb->ip_summed = CHECKSUM_NONE; + skb->csum_not_inet = 0; + + return 1; +} + +static int tcf_csum_ipv4(struct sk_buff *skb, u32 update_flags) +{ + const struct iphdr *iph; + int ntkoff; + + ntkoff = skb_network_offset(skb); + + if (!pskb_may_pull(skb, sizeof(*iph) + ntkoff)) + goto fail; + + iph = ip_hdr(skb); + + switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) { + case IPPROTO_ICMP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_ICMP) + if (!tcf_csum_ipv4_icmp(skb, iph->ihl * 4, + ntohs(iph->tot_len))) + goto fail; + break; + case IPPROTO_IGMP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_IGMP) + if (!tcf_csum_ipv4_igmp(skb, iph->ihl * 4, + ntohs(iph->tot_len))) + goto fail; + break; + case IPPROTO_TCP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_TCP) + if (!tcf_csum_ipv4_tcp(skb, iph->ihl * 4, + ntohs(iph->tot_len))) + goto fail; + break; + case IPPROTO_UDP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDP) + if (!tcf_csum_ipv4_udp(skb, iph->ihl * 4, + ntohs(iph->tot_len), 0)) + goto fail; + break; + case IPPROTO_UDPLITE: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDPLITE) + if (!tcf_csum_ipv4_udp(skb, iph->ihl * 4, + ntohs(iph->tot_len), 1)) + goto fail; + break; + case IPPROTO_SCTP: + if ((update_flags & TCA_CSUM_UPDATE_FLAG_SCTP) && + !tcf_csum_sctp(skb, iph->ihl * 4, ntohs(iph->tot_len))) + goto fail; + break; + } + + if (update_flags & TCA_CSUM_UPDATE_FLAG_IPV4HDR) { + if (skb_try_make_writable(skb, sizeof(*iph) + ntkoff)) + goto fail; + + ip_send_check(ip_hdr(skb)); + } + + return 1; + +fail: + return 0; +} + +static int tcf_csum_ipv6_hopopts(struct ipv6_opt_hdr *ip6xh, unsigned int ixhl, + unsigned int *pl) +{ + int off, len, optlen; + unsigned char *xh = (void *)ip6xh; + + off = sizeof(*ip6xh); + len = ixhl - off; + + while (len > 1) { + switch (xh[off]) { + case IPV6_TLV_PAD1: + optlen = 1; + break; + case IPV6_TLV_JUMBO: + optlen = xh[off + 1] + 2; + if (optlen != 6 || len < 6 || (off & 3) != 2) + /* wrong jumbo option length/alignment */ + return 0; + *pl = ntohl(*(__be32 *)(xh + off + 2)); + goto done; + default: + optlen = xh[off + 1] + 2; + if (optlen > len) + /* ignore obscure options */ + goto done; + break; + } + off += optlen; + len -= optlen; + } + +done: + return 1; +} + +static int tcf_csum_ipv6(struct sk_buff *skb, u32 update_flags) +{ + struct ipv6hdr *ip6h; + struct ipv6_opt_hdr *ip6xh; + unsigned int hl, ixhl; + unsigned int pl; + int ntkoff; + u8 nexthdr; + + ntkoff = skb_network_offset(skb); + + hl = sizeof(*ip6h); + + if (!pskb_may_pull(skb, hl + ntkoff)) + goto fail; + + ip6h = ipv6_hdr(skb); + + pl = ntohs(ip6h->payload_len); + nexthdr = ip6h->nexthdr; + + do { + switch (nexthdr) { + case NEXTHDR_FRAGMENT: + goto ignore_skb; + case NEXTHDR_ROUTING: + case NEXTHDR_HOP: + case NEXTHDR_DEST: + if (!pskb_may_pull(skb, hl + sizeof(*ip6xh) + ntkoff)) + goto fail; + ip6xh = (void *)(skb_network_header(skb) + hl); + ixhl = ipv6_optlen(ip6xh); + if (!pskb_may_pull(skb, hl + ixhl + ntkoff)) + goto fail; + ip6xh = (void *)(skb_network_header(skb) + hl); + if ((nexthdr == NEXTHDR_HOP) && + !(tcf_csum_ipv6_hopopts(ip6xh, ixhl, &pl))) + goto fail; + nexthdr = ip6xh->nexthdr; + hl += ixhl; + break; + case IPPROTO_ICMPV6: + if (update_flags & TCA_CSUM_UPDATE_FLAG_ICMP) + if (!tcf_csum_ipv6_icmp(skb, + hl, pl + sizeof(*ip6h))) + goto fail; + goto done; + case IPPROTO_TCP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_TCP) + if (!tcf_csum_ipv6_tcp(skb, + hl, pl + sizeof(*ip6h))) + goto fail; + goto done; + case IPPROTO_UDP: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDP) + if (!tcf_csum_ipv6_udp(skb, hl, + pl + sizeof(*ip6h), 0)) + goto fail; + goto done; + case IPPROTO_UDPLITE: + if (update_flags & TCA_CSUM_UPDATE_FLAG_UDPLITE) + if (!tcf_csum_ipv6_udp(skb, hl, + pl + sizeof(*ip6h), 1)) + goto fail; + goto done; + case IPPROTO_SCTP: + if ((update_flags & TCA_CSUM_UPDATE_FLAG_SCTP) && + !tcf_csum_sctp(skb, hl, pl + sizeof(*ip6h))) + goto fail; + goto done; + default: + goto ignore_skb; + } + } while (pskb_may_pull(skb, hl + 1 + ntkoff)); + +done: +ignore_skb: + return 1; + +fail: + return 0; +} + +static int tcf_csum_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_csum *p = to_tcf_csum(a); + bool orig_vlan_tag_present = false; + unsigned int vlan_hdr_count = 0; + struct tcf_csum_params *params; + u32 update_flags; + __be16 protocol; + int action; + + params = rcu_dereference_bh(p->params); + + tcf_lastuse_update(&p->tcf_tm); + bstats_cpu_update(this_cpu_ptr(p->common.cpu_bstats), skb); + + action = READ_ONCE(p->tcf_action); + if (unlikely(action == TC_ACT_SHOT)) + goto drop; + + update_flags = params->update_flags; + protocol = skb_protocol(skb, false); +again: + switch (protocol) { + case cpu_to_be16(ETH_P_IP): + if (!tcf_csum_ipv4(skb, update_flags)) + goto drop; + break; + case cpu_to_be16(ETH_P_IPV6): + if (!tcf_csum_ipv6(skb, update_flags)) + goto drop; + break; + case cpu_to_be16(ETH_P_8021AD): /* fall through */ + case cpu_to_be16(ETH_P_8021Q): + if (skb_vlan_tag_present(skb) && !orig_vlan_tag_present) { + protocol = skb->protocol; + orig_vlan_tag_present = true; + } else { + struct vlan_hdr *vlan = (struct vlan_hdr *)skb->data; + + protocol = vlan->h_vlan_encapsulated_proto; + skb_pull(skb, VLAN_HLEN); + skb_reset_network_header(skb); + vlan_hdr_count++; + } + goto again; + } + +out: + /* Restore the skb for the pulled VLAN tags */ + while (vlan_hdr_count--) { + skb_push(skb, VLAN_HLEN); + skb_reset_network_header(skb); + } + + return action; + +drop: + qstats_drop_inc(this_cpu_ptr(p->common.cpu_qstats)); + action = TC_ACT_SHOT; + goto out; +} + +static int tcf_csum_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_csum *p = to_tcf_csum(a); + struct tcf_csum_params *params; + struct tc_csum opt = { + .index = p->tcf_index, + .refcnt = refcount_read(&p->tcf_refcnt) - ref, + .bindcnt = atomic_read(&p->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&p->tcf_lock); + params = rcu_dereference_protected(p->params, + lockdep_is_held(&p->tcf_lock)); + opt.action = p->tcf_action; + opt.update_flags = params->update_flags; + + if (nla_put(skb, TCA_CSUM_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + tcf_tm_dump(&t, &p->tcf_tm); + if (nla_put_64bit(skb, TCA_CSUM_TM, sizeof(t), &t, TCA_CSUM_PAD)) + goto nla_put_failure; + spin_unlock_bh(&p->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&p->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static void tcf_csum_cleanup(struct tc_action *a) +{ + struct tcf_csum *p = to_tcf_csum(a); + struct tcf_csum_params *params; + + params = rcu_dereference_protected(p->params, 1); + if (params) + kfree_rcu(params, rcu); +} + +static int tcf_csum_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, csum_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_csum_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, csum_net_id); + + return tcf_idr_search(tn, a, index); +} + +static size_t tcf_csum_get_fill_size(const struct tc_action *act) +{ + return nla_total_size(sizeof(struct tc_csum)); +} + +static struct tc_action_ops act_csum_ops = { + .kind = "csum", + .type = TCA_ACT_CSUM, + .owner = THIS_MODULE, + .act = tcf_csum_act, + .dump = tcf_csum_dump, + .init = tcf_csum_init, + .cleanup = tcf_csum_cleanup, + .walk = tcf_csum_walker, + .lookup = tcf_csum_search, + .get_fill_size = tcf_csum_get_fill_size, + .size = sizeof(struct tcf_csum), +}; + +static __net_init int csum_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, csum_net_id); + + return tc_action_net_init(net, tn, &act_csum_ops); +} + +static void __net_exit csum_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, csum_net_id); +} + +static struct pernet_operations csum_net_ops = { + .init = csum_init_net, + .exit_batch = csum_exit_net, + .id = &csum_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_DESCRIPTION("Checksum updating actions"); +MODULE_LICENSE("GPL"); + +static int __init csum_init_module(void) +{ + return tcf_register_action(&act_csum_ops, &csum_net_ops); +} + +static void __exit csum_cleanup_module(void) +{ + tcf_unregister_action(&act_csum_ops, &csum_net_ops); +} + +module_init(csum_init_module); +module_exit(csum_cleanup_module); diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c new file mode 100644 index 000000000..dfef96213 --- /dev/null +++ b/net/sched/act_gact.c @@ -0,0 +1,302 @@ +/* + * net/sched/act_gact.c Generic actions + * + * 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. + * + * copyright Jamal Hadi Salim (2002-4) + * + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <linux/tc_act/tc_gact.h> +#include <net/tc_act/tc_gact.h> + +static unsigned int gact_net_id; +static struct tc_action_ops act_gact_ops; + +#ifdef CONFIG_GACT_PROB +static int gact_net_rand(struct tcf_gact *gact) +{ + smp_rmb(); /* coupled with smp_wmb() in tcf_gact_init() */ + if (prandom_u32() % gact->tcfg_pval) + return gact->tcf_action; + return gact->tcfg_paction; +} + +static int gact_determ(struct tcf_gact *gact) +{ + u32 pack = atomic_inc_return(&gact->packets); + + smp_rmb(); /* coupled with smp_wmb() in tcf_gact_init() */ + if (pack % gact->tcfg_pval) + return gact->tcf_action; + return gact->tcfg_paction; +} + +typedef int (*g_rand)(struct tcf_gact *gact); +static g_rand gact_rand[MAX_RAND] = { NULL, gact_net_rand, gact_determ }; +#endif /* CONFIG_GACT_PROB */ + +static const struct nla_policy gact_policy[TCA_GACT_MAX + 1] = { + [TCA_GACT_PARMS] = { .len = sizeof(struct tc_gact) }, + [TCA_GACT_PROB] = { .len = sizeof(struct tc_gact_p) }, +}; + +static int tcf_gact_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, gact_net_id); + struct nlattr *tb[TCA_GACT_MAX + 1]; + struct tc_gact *parm; + struct tcf_gact *gact; + int ret = 0; + u32 index; + int err; +#ifdef CONFIG_GACT_PROB + struct tc_gact_p *p_parm = NULL; +#endif + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_GACT_MAX, nla, gact_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_GACT_PARMS] == NULL) + return -EINVAL; + parm = nla_data(tb[TCA_GACT_PARMS]); + index = parm->index; + +#ifndef CONFIG_GACT_PROB + if (tb[TCA_GACT_PROB] != NULL) + return -EOPNOTSUPP; +#else + if (tb[TCA_GACT_PROB]) { + p_parm = nla_data(tb[TCA_GACT_PROB]); + if (p_parm->ptype >= MAX_RAND) + return -EINVAL; + } +#endif + + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (!err) { + ret = tcf_idr_create(tn, index, est, a, + &act_gact_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else if (err > 0) { + if (bind)/* dont override defaults */ + return 0; + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + } else { + return err; + } + + gact = to_gact(*a); + + spin_lock_bh(&gact->tcf_lock); + gact->tcf_action = parm->action; +#ifdef CONFIG_GACT_PROB + if (p_parm) { + gact->tcfg_paction = p_parm->paction; + gact->tcfg_pval = max_t(u16, 1, p_parm->pval); + /* Make sure tcfg_pval is written before tcfg_ptype + * coupled with smp_rmb() in gact_net_rand() & gact_determ() + */ + smp_wmb(); + gact->tcfg_ptype = p_parm->ptype; + } +#endif + spin_unlock_bh(&gact->tcf_lock); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; +} + +static int tcf_gact_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_gact *gact = to_gact(a); + int action = READ_ONCE(gact->tcf_action); + +#ifdef CONFIG_GACT_PROB + { + u32 ptype = READ_ONCE(gact->tcfg_ptype); + + if (ptype) + action = gact_rand[ptype](gact); + } +#endif + bstats_cpu_update(this_cpu_ptr(gact->common.cpu_bstats), skb); + if (action == TC_ACT_SHOT) + qstats_drop_inc(this_cpu_ptr(gact->common.cpu_qstats)); + + tcf_lastuse_update(&gact->tcf_tm); + + return action; +} + +static void tcf_gact_stats_update(struct tc_action *a, u64 bytes, u32 packets, + u64 lastuse) +{ + struct tcf_gact *gact = to_gact(a); + int action = READ_ONCE(gact->tcf_action); + struct tcf_t *tm = &gact->tcf_tm; + + _bstats_cpu_update(this_cpu_ptr(gact->common.cpu_bstats), bytes, + packets); + if (action == TC_ACT_SHOT) + this_cpu_ptr(gact->common.cpu_qstats)->drops += packets; + + tm->lastuse = max_t(u64, tm->lastuse, lastuse); +} + +static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_gact *gact = to_gact(a); + struct tc_gact opt = { + .index = gact->tcf_index, + .refcnt = refcount_read(&gact->tcf_refcnt) - ref, + .bindcnt = atomic_read(&gact->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&gact->tcf_lock); + opt.action = gact->tcf_action; + if (nla_put(skb, TCA_GACT_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; +#ifdef CONFIG_GACT_PROB + if (gact->tcfg_ptype) { + struct tc_gact_p p_opt = { + .paction = gact->tcfg_paction, + .pval = gact->tcfg_pval, + .ptype = gact->tcfg_ptype, + }; + + if (nla_put(skb, TCA_GACT_PROB, sizeof(p_opt), &p_opt)) + goto nla_put_failure; + } +#endif + tcf_tm_dump(&t, &gact->tcf_tm); + if (nla_put_64bit(skb, TCA_GACT_TM, sizeof(t), &t, TCA_GACT_PAD)) + goto nla_put_failure; + spin_unlock_bh(&gact->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&gact->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_gact_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, gact_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_gact_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, gact_net_id); + + return tcf_idr_search(tn, a, index); +} + +static size_t tcf_gact_get_fill_size(const struct tc_action *act) +{ + size_t sz = nla_total_size(sizeof(struct tc_gact)); /* TCA_GACT_PARMS */ + +#ifdef CONFIG_GACT_PROB + if (to_gact(act)->tcfg_ptype) + /* TCA_GACT_PROB */ + sz += nla_total_size(sizeof(struct tc_gact_p)); +#endif + + return sz; +} + +static struct tc_action_ops act_gact_ops = { + .kind = "gact", + .type = TCA_ACT_GACT, + .owner = THIS_MODULE, + .act = tcf_gact_act, + .stats_update = tcf_gact_stats_update, + .dump = tcf_gact_dump, + .init = tcf_gact_init, + .walk = tcf_gact_walker, + .lookup = tcf_gact_search, + .get_fill_size = tcf_gact_get_fill_size, + .size = sizeof(struct tcf_gact), +}; + +static __net_init int gact_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, gact_net_id); + + return tc_action_net_init(net, tn, &act_gact_ops); +} + +static void __net_exit gact_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, gact_net_id); +} + +static struct pernet_operations gact_net_ops = { + .init = gact_init_net, + .exit_batch = gact_exit_net, + .id = &gact_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2002-4)"); +MODULE_DESCRIPTION("Generic Classifier actions"); +MODULE_LICENSE("GPL"); + +static int __init gact_init_module(void) +{ +#ifdef CONFIG_GACT_PROB + pr_info("GACT probability on\n"); +#else + pr_info("GACT probability NOT on\n"); +#endif + + return tcf_register_action(&act_gact_ops, &gact_net_ops); +} + +static void __exit gact_cleanup_module(void) +{ + tcf_unregister_action(&act_gact_ops, &gact_net_ops); +} + +module_init(gact_init_module); +module_exit(gact_cleanup_module); diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c new file mode 100644 index 000000000..bac353bea --- /dev/null +++ b/net/sched/act_ife.c @@ -0,0 +1,920 @@ +/* + * net/sched/ife.c Inter-FE action based on ForCES WG InterFE LFB + * + * Refer to: + * draft-ietf-forces-interfelfb-03 + * and + * netdev01 paper: + * "Distributing Linux Traffic Control Classifier-Action + * Subsystem" + * Authors: Jamal Hadi Salim and Damascene M. Joachimpillai + * + * 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. + * + * copyright Jamal Hadi Salim (2015) + * +*/ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <net/net_namespace.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <uapi/linux/tc_act/tc_ife.h> +#include <net/tc_act/tc_ife.h> +#include <linux/etherdevice.h> +#include <net/ife.h> + +static unsigned int ife_net_id; +static int max_metacnt = IFE_META_MAX + 1; +static struct tc_action_ops act_ife_ops; + +static const struct nla_policy ife_policy[TCA_IFE_MAX + 1] = { + [TCA_IFE_PARMS] = { .len = sizeof(struct tc_ife)}, + [TCA_IFE_DMAC] = { .len = ETH_ALEN}, + [TCA_IFE_SMAC] = { .len = ETH_ALEN}, + [TCA_IFE_TYPE] = { .type = NLA_U16}, +}; + +int ife_encode_meta_u16(u16 metaval, void *skbdata, struct tcf_meta_info *mi) +{ + u16 edata = 0; + + if (mi->metaval) + edata = *(u16 *)mi->metaval; + else if (metaval) + edata = metaval; + + if (!edata) /* will not encode */ + return 0; + + edata = htons(edata); + return ife_tlv_meta_encode(skbdata, mi->metaid, 2, &edata); +} +EXPORT_SYMBOL_GPL(ife_encode_meta_u16); + +int ife_get_meta_u32(struct sk_buff *skb, struct tcf_meta_info *mi) +{ + if (mi->metaval) + return nla_put_u32(skb, mi->metaid, *(u32 *)mi->metaval); + else + return nla_put(skb, mi->metaid, 0, NULL); +} +EXPORT_SYMBOL_GPL(ife_get_meta_u32); + +int ife_check_meta_u32(u32 metaval, struct tcf_meta_info *mi) +{ + if (metaval || mi->metaval) + return 8; /* T+L+V == 2+2+4 */ + + return 0; +} +EXPORT_SYMBOL_GPL(ife_check_meta_u32); + +int ife_check_meta_u16(u16 metaval, struct tcf_meta_info *mi) +{ + if (metaval || mi->metaval) + return 8; /* T+L+(V) == 2+2+(2+2bytepad) */ + + return 0; +} +EXPORT_SYMBOL_GPL(ife_check_meta_u16); + +int ife_encode_meta_u32(u32 metaval, void *skbdata, struct tcf_meta_info *mi) +{ + u32 edata = metaval; + + if (mi->metaval) + edata = *(u32 *)mi->metaval; + else if (metaval) + edata = metaval; + + if (!edata) /* will not encode */ + return 0; + + edata = htonl(edata); + return ife_tlv_meta_encode(skbdata, mi->metaid, 4, &edata); +} +EXPORT_SYMBOL_GPL(ife_encode_meta_u32); + +int ife_get_meta_u16(struct sk_buff *skb, struct tcf_meta_info *mi) +{ + if (mi->metaval) + return nla_put_u16(skb, mi->metaid, *(u16 *)mi->metaval); + else + return nla_put(skb, mi->metaid, 0, NULL); +} +EXPORT_SYMBOL_GPL(ife_get_meta_u16); + +int ife_alloc_meta_u32(struct tcf_meta_info *mi, void *metaval, gfp_t gfp) +{ + mi->metaval = kmemdup(metaval, sizeof(u32), gfp); + if (!mi->metaval) + return -ENOMEM; + + return 0; +} +EXPORT_SYMBOL_GPL(ife_alloc_meta_u32); + +int ife_alloc_meta_u16(struct tcf_meta_info *mi, void *metaval, gfp_t gfp) +{ + mi->metaval = kmemdup(metaval, sizeof(u16), gfp); + if (!mi->metaval) + return -ENOMEM; + + return 0; +} +EXPORT_SYMBOL_GPL(ife_alloc_meta_u16); + +void ife_release_meta_gen(struct tcf_meta_info *mi) +{ + kfree(mi->metaval); +} +EXPORT_SYMBOL_GPL(ife_release_meta_gen); + +int ife_validate_meta_u32(void *val, int len) +{ + if (len == sizeof(u32)) + return 0; + + return -EINVAL; +} +EXPORT_SYMBOL_GPL(ife_validate_meta_u32); + +int ife_validate_meta_u16(void *val, int len) +{ + /* length will not include padding */ + if (len == sizeof(u16)) + return 0; + + return -EINVAL; +} +EXPORT_SYMBOL_GPL(ife_validate_meta_u16); + +static LIST_HEAD(ifeoplist); +static DEFINE_RWLOCK(ife_mod_lock); + +static struct tcf_meta_ops *find_ife_oplist(u16 metaid) +{ + struct tcf_meta_ops *o; + + read_lock(&ife_mod_lock); + list_for_each_entry(o, &ifeoplist, list) { + if (o->metaid == metaid) { + if (!try_module_get(o->owner)) + o = NULL; + read_unlock(&ife_mod_lock); + return o; + } + } + read_unlock(&ife_mod_lock); + + return NULL; +} + +int register_ife_op(struct tcf_meta_ops *mops) +{ + struct tcf_meta_ops *m; + + if (!mops->metaid || !mops->metatype || !mops->name || + !mops->check_presence || !mops->encode || !mops->decode || + !mops->get || !mops->alloc) + return -EINVAL; + + write_lock(&ife_mod_lock); + + list_for_each_entry(m, &ifeoplist, list) { + if (m->metaid == mops->metaid || + (strcmp(mops->name, m->name) == 0)) { + write_unlock(&ife_mod_lock); + return -EEXIST; + } + } + + if (!mops->release) + mops->release = ife_release_meta_gen; + + list_add_tail(&mops->list, &ifeoplist); + write_unlock(&ife_mod_lock); + return 0; +} +EXPORT_SYMBOL_GPL(unregister_ife_op); + +int unregister_ife_op(struct tcf_meta_ops *mops) +{ + struct tcf_meta_ops *m; + int err = -ENOENT; + + write_lock(&ife_mod_lock); + list_for_each_entry(m, &ifeoplist, list) { + if (m->metaid == mops->metaid) { + list_del(&mops->list); + err = 0; + break; + } + } + write_unlock(&ife_mod_lock); + + return err; +} +EXPORT_SYMBOL_GPL(register_ife_op); + +static int ife_validate_metatype(struct tcf_meta_ops *ops, void *val, int len) +{ + int ret = 0; + /* XXX: unfortunately cant use nla_policy at this point + * because a length of 0 is valid in the case of + * "allow". "use" semantics do enforce for proper + * length and i couldve use nla_policy but it makes it hard + * to use it just for that.. + */ + if (ops->validate) + return ops->validate(val, len); + + if (ops->metatype == NLA_U32) + ret = ife_validate_meta_u32(val, len); + else if (ops->metatype == NLA_U16) + ret = ife_validate_meta_u16(val, len); + + return ret; +} + +#ifdef CONFIG_MODULES +static const char *ife_meta_id2name(u32 metaid) +{ + switch (metaid) { + case IFE_META_SKBMARK: + return "skbmark"; + case IFE_META_PRIO: + return "skbprio"; + case IFE_META_TCINDEX: + return "tcindex"; + default: + return "unknown"; + } +} +#endif + +/* called when adding new meta information +*/ +static int load_metaops_and_vet(u32 metaid, void *val, int len, bool rtnl_held) +{ + struct tcf_meta_ops *ops = find_ife_oplist(metaid); + int ret = 0; + + if (!ops) { + ret = -ENOENT; +#ifdef CONFIG_MODULES + if (rtnl_held) + rtnl_unlock(); + request_module("ife-meta-%s", ife_meta_id2name(metaid)); + if (rtnl_held) + rtnl_lock(); + ops = find_ife_oplist(metaid); +#endif + } + + if (ops) { + ret = 0; + if (len) + ret = ife_validate_metatype(ops, val, len); + + module_put(ops->owner); + } + + return ret; +} + +/* called when adding new meta information +*/ +static int __add_metainfo(const struct tcf_meta_ops *ops, + struct tcf_ife_info *ife, u32 metaid, void *metaval, + int len, bool atomic, bool exists) +{ + struct tcf_meta_info *mi = NULL; + int ret = 0; + + mi = kzalloc(sizeof(*mi), atomic ? GFP_ATOMIC : GFP_KERNEL); + if (!mi) + return -ENOMEM; + + mi->metaid = metaid; + mi->ops = ops; + if (len > 0) { + ret = ops->alloc(mi, metaval, atomic ? GFP_ATOMIC : GFP_KERNEL); + if (ret != 0) { + kfree(mi); + return ret; + } + } + + if (exists) + spin_lock_bh(&ife->tcf_lock); + list_add_tail(&mi->metalist, &ife->metalist); + if (exists) + spin_unlock_bh(&ife->tcf_lock); + + return ret; +} + +static int add_metainfo_and_get_ops(const struct tcf_meta_ops *ops, + struct tcf_ife_info *ife, u32 metaid, + bool exists) +{ + int ret; + + if (!try_module_get(ops->owner)) + return -ENOENT; + ret = __add_metainfo(ops, ife, metaid, NULL, 0, true, exists); + if (ret) + module_put(ops->owner); + return ret; +} + +static int add_metainfo(struct tcf_ife_info *ife, u32 metaid, void *metaval, + int len, bool exists) +{ + const struct tcf_meta_ops *ops = find_ife_oplist(metaid); + int ret; + + if (!ops) + return -ENOENT; + ret = __add_metainfo(ops, ife, metaid, metaval, len, false, exists); + if (ret) + /*put back what find_ife_oplist took */ + module_put(ops->owner); + return ret; +} + +static int use_all_metadata(struct tcf_ife_info *ife, bool exists) +{ + struct tcf_meta_ops *o; + int rc = 0; + int installed = 0; + + read_lock(&ife_mod_lock); + list_for_each_entry(o, &ifeoplist, list) { + rc = add_metainfo_and_get_ops(o, ife, o->metaid, exists); + if (rc == 0) + installed += 1; + } + read_unlock(&ife_mod_lock); + + if (installed) + return 0; + else + return -EINVAL; +} + +static int dump_metalist(struct sk_buff *skb, struct tcf_ife_info *ife) +{ + struct tcf_meta_info *e; + struct nlattr *nest; + unsigned char *b = skb_tail_pointer(skb); + int total_encoded = 0; + + /*can only happen on decode */ + if (list_empty(&ife->metalist)) + return 0; + + nest = nla_nest_start(skb, TCA_IFE_METALST); + if (!nest) + goto out_nlmsg_trim; + + list_for_each_entry(e, &ife->metalist, metalist) { + if (!e->ops->get(skb, e)) + total_encoded += 1; + } + + if (!total_encoded) + goto out_nlmsg_trim; + + nla_nest_end(skb, nest); + + return 0; + +out_nlmsg_trim: + nlmsg_trim(skb, b); + return -1; +} + +/* under ife->tcf_lock */ +static void _tcf_ife_cleanup(struct tc_action *a) +{ + struct tcf_ife_info *ife = to_ife(a); + struct tcf_meta_info *e, *n; + + list_for_each_entry_safe(e, n, &ife->metalist, metalist) { + list_del(&e->metalist); + if (e->metaval) { + if (e->ops->release) + e->ops->release(e); + else + kfree(e->metaval); + } + module_put(e->ops->owner); + kfree(e); + } +} + +static void tcf_ife_cleanup(struct tc_action *a) +{ + struct tcf_ife_info *ife = to_ife(a); + struct tcf_ife_params *p; + + spin_lock_bh(&ife->tcf_lock); + _tcf_ife_cleanup(a); + spin_unlock_bh(&ife->tcf_lock); + + p = rcu_dereference_protected(ife->params, 1); + if (p) + kfree_rcu(p, rcu); +} + +static int populate_metalist(struct tcf_ife_info *ife, struct nlattr **tb, + bool exists, bool rtnl_held) +{ + int len = 0; + int rc = 0; + int i = 0; + void *val; + + for (i = 1; i < max_metacnt; i++) { + if (tb[i]) { + val = nla_data(tb[i]); + len = nla_len(tb[i]); + + rc = load_metaops_and_vet(i, val, len, rtnl_held); + if (rc != 0) + return rc; + + rc = add_metainfo(ife, i, val, len, exists); + if (rc) + return rc; + } + } + + return rc; +} + +static int tcf_ife_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, ife_net_id); + struct nlattr *tb[TCA_IFE_MAX + 1]; + struct nlattr *tb2[IFE_META_MAX + 1]; + struct tcf_ife_params *p; + struct tcf_ife_info *ife; + u16 ife_type = ETH_P_IFE; + struct tc_ife *parm; + u8 *daddr = NULL; + u8 *saddr = NULL; + bool exists = false; + int ret = 0; + u32 index; + int err; + + if (!nla) { + NL_SET_ERR_MSG_MOD(extack, "IFE requires attributes to be passed"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_IFE_MAX, nla, ife_policy, NULL); + if (err < 0) + return err; + + if (!tb[TCA_IFE_PARMS]) + return -EINVAL; + + parm = nla_data(tb[TCA_IFE_PARMS]); + + /* IFE_DECODE is 0 and indicates the opposite of IFE_ENCODE because + * they cannot run as the same time. Check on all other values which + * are not supported right now. + */ + if (parm->flags & ~IFE_ENCODE) + return -EINVAL; + + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) + return -ENOMEM; + + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) { + kfree(p); + return err; + } + exists = err; + if (exists && bind) { + kfree(p); + return 0; + } + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, &act_ife_ops, + bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + kfree(p); + return ret; + } + ret = ACT_P_CREATED; + } else if (!ovr) { + tcf_idr_release(*a, bind); + kfree(p); + return -EEXIST; + } + + ife = to_ife(*a); + p->flags = parm->flags; + + if (parm->flags & IFE_ENCODE) { + if (tb[TCA_IFE_TYPE]) + ife_type = nla_get_u16(tb[TCA_IFE_TYPE]); + if (tb[TCA_IFE_DMAC]) + daddr = nla_data(tb[TCA_IFE_DMAC]); + if (tb[TCA_IFE_SMAC]) + saddr = nla_data(tb[TCA_IFE_SMAC]); + } + + if (parm->flags & IFE_ENCODE) { + if (daddr) + ether_addr_copy(p->eth_dst, daddr); + else + eth_zero_addr(p->eth_dst); + + if (saddr) + ether_addr_copy(p->eth_src, saddr); + else + eth_zero_addr(p->eth_src); + + p->eth_type = ife_type; + } + + + if (ret == ACT_P_CREATED) + INIT_LIST_HEAD(&ife->metalist); + + if (tb[TCA_IFE_METALST]) { + err = nla_parse_nested(tb2, IFE_META_MAX, tb[TCA_IFE_METALST], + NULL, NULL); + if (err) { +metadata_parse_err: + tcf_idr_release(*a, bind); + kfree(p); + return err; + } + + err = populate_metalist(ife, tb2, exists, rtnl_held); + if (err) + goto metadata_parse_err; + + } else { + /* if no passed metadata allow list or passed allow-all + * then here we process by adding as many supported metadatum + * as we can. You better have at least one else we are + * going to bail out + */ + err = use_all_metadata(ife, exists); + if (err) { + tcf_idr_release(*a, bind); + kfree(p); + return err; + } + } + + if (exists) + spin_lock_bh(&ife->tcf_lock); + ife->tcf_action = parm->action; + /* protected by tcf_lock when modifying existing action */ + rcu_swap_protected(ife->params, p, 1); + + if (exists) + spin_unlock_bh(&ife->tcf_lock); + if (p) + kfree_rcu(p, rcu); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + + return ret; +} + +static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_ife_info *ife = to_ife(a); + struct tcf_ife_params *p; + struct tc_ife opt = { + .index = ife->tcf_index, + .refcnt = refcount_read(&ife->tcf_refcnt) - ref, + .bindcnt = atomic_read(&ife->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&ife->tcf_lock); + opt.action = ife->tcf_action; + p = rcu_dereference_protected(ife->params, + lockdep_is_held(&ife->tcf_lock)); + opt.flags = p->flags; + + if (nla_put(skb, TCA_IFE_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + tcf_tm_dump(&t, &ife->tcf_tm); + if (nla_put_64bit(skb, TCA_IFE_TM, sizeof(t), &t, TCA_IFE_PAD)) + goto nla_put_failure; + + if (!is_zero_ether_addr(p->eth_dst)) { + if (nla_put(skb, TCA_IFE_DMAC, ETH_ALEN, p->eth_dst)) + goto nla_put_failure; + } + + if (!is_zero_ether_addr(p->eth_src)) { + if (nla_put(skb, TCA_IFE_SMAC, ETH_ALEN, p->eth_src)) + goto nla_put_failure; + } + + if (nla_put(skb, TCA_IFE_TYPE, 2, &p->eth_type)) + goto nla_put_failure; + + if (dump_metalist(skb, ife)) { + /*ignore failure to dump metalist */ + pr_info("Failed to dump metalist\n"); + } + + spin_unlock_bh(&ife->tcf_lock); + return skb->len; + +nla_put_failure: + spin_unlock_bh(&ife->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int find_decode_metaid(struct sk_buff *skb, struct tcf_ife_info *ife, + u16 metaid, u16 mlen, void *mdata) +{ + struct tcf_meta_info *e; + + /* XXX: use hash to speed up */ + list_for_each_entry(e, &ife->metalist, metalist) { + if (metaid == e->metaid) { + if (e->ops) { + /* We check for decode presence already */ + return e->ops->decode(skb, mdata, mlen); + } + } + } + + return -ENOENT; +} + +static int tcf_ife_decode(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_ife_info *ife = to_ife(a); + int action = ife->tcf_action; + u8 *ifehdr_end; + u8 *tlv_data; + u16 metalen; + + bstats_cpu_update(this_cpu_ptr(ife->common.cpu_bstats), skb); + tcf_lastuse_update(&ife->tcf_tm); + + if (skb_at_tc_ingress(skb)) + skb_push(skb, skb->dev->hard_header_len); + + tlv_data = ife_decode(skb, &metalen); + if (unlikely(!tlv_data)) { + qstats_drop_inc(this_cpu_ptr(ife->common.cpu_qstats)); + return TC_ACT_SHOT; + } + + ifehdr_end = tlv_data + metalen; + for (; tlv_data < ifehdr_end; tlv_data = ife_tlv_meta_next(tlv_data)) { + u8 *curr_data; + u16 mtype; + u16 dlen; + + curr_data = ife_tlv_meta_decode(tlv_data, ifehdr_end, &mtype, + &dlen, NULL); + if (!curr_data) { + qstats_drop_inc(this_cpu_ptr(ife->common.cpu_qstats)); + return TC_ACT_SHOT; + } + + if (find_decode_metaid(skb, ife, mtype, dlen, curr_data)) { + /* abuse overlimits to count when we receive metadata + * but dont have an ops for it + */ + pr_info_ratelimited("Unknown metaid %d dlen %d\n", + mtype, dlen); + qstats_overlimit_inc(this_cpu_ptr(ife->common.cpu_qstats)); + } + } + + if (WARN_ON(tlv_data != ifehdr_end)) { + qstats_drop_inc(this_cpu_ptr(ife->common.cpu_qstats)); + return TC_ACT_SHOT; + } + + skb->protocol = eth_type_trans(skb, skb->dev); + skb_reset_network_header(skb); + + return action; +} + +/*XXX: check if we can do this at install time instead of current + * send data path +**/ +static int ife_get_sz(struct sk_buff *skb, struct tcf_ife_info *ife) +{ + struct tcf_meta_info *e, *n; + int tot_run_sz = 0, run_sz = 0; + + list_for_each_entry_safe(e, n, &ife->metalist, metalist) { + if (e->ops->check_presence) { + run_sz = e->ops->check_presence(skb, e); + tot_run_sz += run_sz; + } + } + + return tot_run_sz; +} + +static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res, struct tcf_ife_params *p) +{ + struct tcf_ife_info *ife = to_ife(a); + int action = ife->tcf_action; + struct ethhdr *oethh; /* outer ether header */ + struct tcf_meta_info *e; + /* + OUTERHDR:TOTMETALEN:{TLVHDR:Metadatum:TLVHDR..}:ORIGDATA + where ORIGDATA = original ethernet header ... + */ + u16 metalen = ife_get_sz(skb, ife); + int hdrm = metalen + skb->dev->hard_header_len + IFE_METAHDRLEN; + unsigned int skboff = 0; + int new_len = skb->len + hdrm; + bool exceed_mtu = false; + void *ife_meta; + int err = 0; + + if (!skb_at_tc_ingress(skb)) { + if (new_len > skb->dev->mtu) + exceed_mtu = true; + } + + bstats_cpu_update(this_cpu_ptr(ife->common.cpu_bstats), skb); + tcf_lastuse_update(&ife->tcf_tm); + + if (!metalen) { /* no metadata to send */ + /* abuse overlimits to count when we allow packet + * with no metadata + */ + qstats_overlimit_inc(this_cpu_ptr(ife->common.cpu_qstats)); + return action; + } + /* could be stupid policy setup or mtu config + * so lets be conservative.. */ + if ((action == TC_ACT_SHOT) || exceed_mtu) { + qstats_drop_inc(this_cpu_ptr(ife->common.cpu_qstats)); + return TC_ACT_SHOT; + } + + if (skb_at_tc_ingress(skb)) + skb_push(skb, skb->dev->hard_header_len); + + ife_meta = ife_encode(skb, metalen); + + spin_lock(&ife->tcf_lock); + + /* XXX: we dont have a clever way of telling encode to + * not repeat some of the computations that are done by + * ops->presence_check... + */ + list_for_each_entry(e, &ife->metalist, metalist) { + if (e->ops->encode) { + err = e->ops->encode(skb, (void *)(ife_meta + skboff), + e); + } + if (err < 0) { + /* too corrupt to keep around if overwritten */ + spin_unlock(&ife->tcf_lock); + qstats_drop_inc(this_cpu_ptr(ife->common.cpu_qstats)); + return TC_ACT_SHOT; + } + skboff += err; + } + spin_unlock(&ife->tcf_lock); + oethh = (struct ethhdr *)skb->data; + + if (!is_zero_ether_addr(p->eth_src)) + ether_addr_copy(oethh->h_source, p->eth_src); + if (!is_zero_ether_addr(p->eth_dst)) + ether_addr_copy(oethh->h_dest, p->eth_dst); + oethh->h_proto = htons(p->eth_type); + + if (skb_at_tc_ingress(skb)) + skb_pull(skb, skb->dev->hard_header_len); + + return action; +} + +static int tcf_ife_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_ife_info *ife = to_ife(a); + struct tcf_ife_params *p; + int ret; + + p = rcu_dereference_bh(ife->params); + if (p->flags & IFE_ENCODE) { + ret = tcf_ife_encode(skb, a, res, p); + return ret; + } + + return tcf_ife_decode(skb, a, res); +} + +static int tcf_ife_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, ife_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_ife_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, ife_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_ife_ops = { + .kind = "ife", + .type = TCA_ACT_IFE, + .owner = THIS_MODULE, + .act = tcf_ife_act, + .dump = tcf_ife_dump, + .cleanup = tcf_ife_cleanup, + .init = tcf_ife_init, + .walk = tcf_ife_walker, + .lookup = tcf_ife_search, + .size = sizeof(struct tcf_ife_info), +}; + +static __net_init int ife_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, ife_net_id); + + return tc_action_net_init(net, tn, &act_ife_ops); +} + +static void __net_exit ife_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, ife_net_id); +} + +static struct pernet_operations ife_net_ops = { + .init = ife_init_net, + .exit_batch = ife_exit_net, + .id = &ife_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init ife_init_module(void) +{ + return tcf_register_action(&act_ife_ops, &ife_net_ops); +} + +static void __exit ife_cleanup_module(void) +{ + tcf_unregister_action(&act_ife_ops, &ife_net_ops); +} + +module_init(ife_init_module); +module_exit(ife_cleanup_module); + +MODULE_AUTHOR("Jamal Hadi Salim(2015)"); +MODULE_DESCRIPTION("Inter-FE LFB action"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c new file mode 100644 index 000000000..01d3669ef --- /dev/null +++ b/net/sched/act_ipt.c @@ -0,0 +1,452 @@ +/* + * net/sched/act_ipt.c iptables target interface + * + *TODO: Add other tables. For now we only support the ipv4 table targets + * + * 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. + * + * Copyright: Jamal Hadi Salim (2002-13) + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <linux/tc_act/tc_ipt.h> +#include <net/tc_act/tc_ipt.h> + +#include <linux/netfilter_ipv4/ip_tables.h> + + +static unsigned int ipt_net_id; +static struct tc_action_ops act_ipt_ops; + +static unsigned int xt_net_id; +static struct tc_action_ops act_xt_ops; + +static int ipt_init_target(struct net *net, struct xt_entry_target *t, + char *table, unsigned int hook) +{ + struct xt_tgchk_param par; + struct xt_target *target; + struct ipt_entry e = {}; + int ret = 0; + + target = xt_request_find_target(AF_INET, t->u.user.name, + t->u.user.revision); + if (IS_ERR(target)) + return PTR_ERR(target); + + t->u.kernel.target = target; + memset(&par, 0, sizeof(par)); + par.net = net; + par.table = table; + par.entryinfo = &e; + par.target = target; + par.targinfo = t->data; + par.hook_mask = hook; + par.family = NFPROTO_IPV4; + + ret = xt_check_target(&par, t->u.target_size - sizeof(*t), 0, false); + if (ret < 0) { + module_put(t->u.kernel.target->me); + return ret; + } + return 0; +} + +static void ipt_destroy_target(struct xt_entry_target *t, struct net *net) +{ + struct xt_tgdtor_param par = { + .target = t->u.kernel.target, + .targinfo = t->data, + .family = NFPROTO_IPV4, + .net = net, + }; + if (par.target->destroy != NULL) + par.target->destroy(&par); + module_put(par.target->me); +} + +static void tcf_ipt_release(struct tc_action *a) +{ + struct tcf_ipt *ipt = to_ipt(a); + + if (ipt->tcfi_t) { + ipt_destroy_target(ipt->tcfi_t, a->idrinfo->net); + kfree(ipt->tcfi_t); + } + kfree(ipt->tcfi_tname); +} + +static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = { + [TCA_IPT_TABLE] = { .type = NLA_STRING, .len = IFNAMSIZ }, + [TCA_IPT_HOOK] = { .type = NLA_U32 }, + [TCA_IPT_INDEX] = { .type = NLA_U32 }, + [TCA_IPT_TARG] = { .len = sizeof(struct xt_entry_target) }, +}; + +static int __tcf_ipt_init(struct net *net, unsigned int id, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + const struct tc_action_ops *ops, int ovr, int bind) +{ + struct tc_action_net *tn = net_generic(net, id); + struct nlattr *tb[TCA_IPT_MAX + 1]; + struct tcf_ipt *ipt; + struct xt_entry_target *td, *t; + char *tname; + bool exists = false; + int ret = 0, err; + u32 hook = 0; + u32 index = 0; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_IPT_MAX, nla, ipt_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_IPT_INDEX] != NULL) + index = nla_get_u32(tb[TCA_IPT_INDEX]); + + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + if (tb[TCA_IPT_HOOK] == NULL || tb[TCA_IPT_TARG] == NULL) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + + td = (struct xt_entry_target *)nla_data(tb[TCA_IPT_TARG]); + if (nla_len(tb[TCA_IPT_TARG]) != td->u.target_size) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, ops, bind, + false); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else { + if (bind)/* dont override defaults */ + return 0; + + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + } + hook = nla_get_u32(tb[TCA_IPT_HOOK]); + + err = -ENOMEM; + tname = kmalloc(IFNAMSIZ, GFP_KERNEL); + if (unlikely(!tname)) + goto err1; + if (tb[TCA_IPT_TABLE] == NULL || + nla_strlcpy(tname, tb[TCA_IPT_TABLE], IFNAMSIZ) >= IFNAMSIZ) + strcpy(tname, "mangle"); + + t = kmemdup(td, td->u.target_size, GFP_KERNEL); + if (unlikely(!t)) + goto err2; + + err = ipt_init_target(net, t, tname, hook); + if (err < 0) + goto err3; + + ipt = to_ipt(*a); + + spin_lock_bh(&ipt->tcf_lock); + if (ret != ACT_P_CREATED) { + ipt_destroy_target(ipt->tcfi_t, net); + kfree(ipt->tcfi_tname); + kfree(ipt->tcfi_t); + } + ipt->tcfi_tname = tname; + ipt->tcfi_t = t; + ipt->tcfi_hook = hook; + spin_unlock_bh(&ipt->tcf_lock); + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; + +err3: + kfree(t); +err2: + kfree(tname); +err1: + tcf_idr_release(*a, bind); + return err; +} + +static int tcf_ipt_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, int ovr, + int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + return __tcf_ipt_init(net, ipt_net_id, nla, est, a, &act_ipt_ops, ovr, + bind); +} + +static int tcf_xt_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, int ovr, + int bind, bool unlocked, + struct netlink_ext_ack *extack) +{ + return __tcf_ipt_init(net, xt_net_id, nla, est, a, &act_xt_ops, ovr, + bind); +} + +static int tcf_ipt_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + int ret = 0, result = 0; + struct tcf_ipt *ipt = to_ipt(a); + struct xt_action_param par; + struct nf_hook_state state = { + .net = dev_net(skb->dev), + .in = skb->dev, + .hook = ipt->tcfi_hook, + .pf = NFPROTO_IPV4, + }; + + if (skb_unclone(skb, GFP_ATOMIC)) + return TC_ACT_UNSPEC; + + spin_lock(&ipt->tcf_lock); + + tcf_lastuse_update(&ipt->tcf_tm); + bstats_update(&ipt->tcf_bstats, skb); + + /* yes, we have to worry about both in and out dev + * worry later - danger - this API seems to have changed + * from earlier kernels + */ + par.state = &state; + par.target = ipt->tcfi_t->u.kernel.target; + par.targinfo = ipt->tcfi_t->data; + ret = par.target->target(skb, &par); + + switch (ret) { + case NF_ACCEPT: + result = TC_ACT_OK; + break; + case NF_DROP: + result = TC_ACT_SHOT; + ipt->tcf_qstats.drops++; + break; + case XT_CONTINUE: + result = TC_ACT_PIPE; + break; + default: + net_notice_ratelimited("tc filter: Bogus netfilter code %d assume ACCEPT\n", + ret); + result = TC_ACT_OK; + break; + } + spin_unlock(&ipt->tcf_lock); + return result; + +} + +static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_ipt *ipt = to_ipt(a); + struct xt_entry_target *t; + struct tcf_t tm; + struct tc_cnt c; + + /* for simple targets kernel size == user size + * user name = target name + * for foolproof you need to not assume this + */ + + spin_lock_bh(&ipt->tcf_lock); + t = kmemdup(ipt->tcfi_t, ipt->tcfi_t->u.user.target_size, GFP_ATOMIC); + if (unlikely(!t)) + goto nla_put_failure; + + c.bindcnt = atomic_read(&ipt->tcf_bindcnt) - bind; + c.refcnt = refcount_read(&ipt->tcf_refcnt) - ref; + strcpy(t->u.user.name, ipt->tcfi_t->u.kernel.target->name); + + if (nla_put(skb, TCA_IPT_TARG, ipt->tcfi_t->u.user.target_size, t) || + nla_put_u32(skb, TCA_IPT_INDEX, ipt->tcf_index) || + nla_put_u32(skb, TCA_IPT_HOOK, ipt->tcfi_hook) || + nla_put(skb, TCA_IPT_CNT, sizeof(struct tc_cnt), &c) || + nla_put_string(skb, TCA_IPT_TABLE, ipt->tcfi_tname)) + goto nla_put_failure; + + tcf_tm_dump(&tm, &ipt->tcf_tm); + if (nla_put_64bit(skb, TCA_IPT_TM, sizeof(tm), &tm, TCA_IPT_PAD)) + goto nla_put_failure; + + spin_unlock_bh(&ipt->tcf_lock); + kfree(t); + return skb->len; + +nla_put_failure: + spin_unlock_bh(&ipt->tcf_lock); + nlmsg_trim(skb, b); + kfree(t); + return -1; +} + +static int tcf_ipt_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, ipt_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_ipt_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, ipt_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_ipt_ops = { + .kind = "ipt", + .type = TCA_ACT_IPT, + .owner = THIS_MODULE, + .act = tcf_ipt_act, + .dump = tcf_ipt_dump, + .cleanup = tcf_ipt_release, + .init = tcf_ipt_init, + .walk = tcf_ipt_walker, + .lookup = tcf_ipt_search, + .size = sizeof(struct tcf_ipt), +}; + +static __net_init int ipt_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, ipt_net_id); + + return tc_action_net_init(net, tn, &act_ipt_ops); +} + +static void __net_exit ipt_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, ipt_net_id); +} + +static struct pernet_operations ipt_net_ops = { + .init = ipt_init_net, + .exit_batch = ipt_exit_net, + .id = &ipt_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int tcf_xt_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, xt_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_xt_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, xt_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_xt_ops = { + .kind = "xt", + .type = TCA_ACT_XT, + .owner = THIS_MODULE, + .act = tcf_ipt_act, + .dump = tcf_ipt_dump, + .cleanup = tcf_ipt_release, + .init = tcf_xt_init, + .walk = tcf_xt_walker, + .lookup = tcf_xt_search, + .size = sizeof(struct tcf_ipt), +}; + +static __net_init int xt_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, xt_net_id); + + return tc_action_net_init(net, tn, &act_xt_ops); +} + +static void __net_exit xt_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, xt_net_id); +} + +static struct pernet_operations xt_net_ops = { + .init = xt_init_net, + .exit_batch = xt_exit_net, + .id = &xt_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2002-13)"); +MODULE_DESCRIPTION("Iptables target actions"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("act_xt"); + +static int __init ipt_init_module(void) +{ + int ret1, ret2; + + ret1 = tcf_register_action(&act_xt_ops, &xt_net_ops); + if (ret1 < 0) + pr_err("Failed to load xt action\n"); + + ret2 = tcf_register_action(&act_ipt_ops, &ipt_net_ops); + if (ret2 < 0) + pr_err("Failed to load ipt action\n"); + + if (ret1 < 0 && ret2 < 0) { + return ret1; + } else + return 0; +} + +static void __exit ipt_cleanup_module(void) +{ + tcf_unregister_action(&act_ipt_ops, &ipt_net_ops); + tcf_unregister_action(&act_xt_ops, &xt_net_ops); +} + +module_init(ipt_init_module); +module_exit(ipt_cleanup_module); diff --git a/net/sched/act_meta_mark.c b/net/sched/act_meta_mark.c new file mode 100644 index 000000000..6445184b2 --- /dev/null +++ b/net/sched/act_meta_mark.c @@ -0,0 +1,78 @@ +/* + * net/sched/act_meta_mark.c IFE skb->mark metadata module + * + * 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. + * + * copyright Jamal Hadi Salim (2015) + * +*/ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <uapi/linux/tc_act/tc_ife.h> +#include <net/tc_act/tc_ife.h> + +static int skbmark_encode(struct sk_buff *skb, void *skbdata, + struct tcf_meta_info *e) +{ + u32 ifemark = skb->mark; + + return ife_encode_meta_u32(ifemark, skbdata, e); +} + +static int skbmark_decode(struct sk_buff *skb, void *data, u16 len) +{ + u32 ifemark = *(u32 *)data; + + skb->mark = ntohl(ifemark); + return 0; +} + +static int skbmark_check(struct sk_buff *skb, struct tcf_meta_info *e) +{ + return ife_check_meta_u32(skb->mark, e); +} + +static struct tcf_meta_ops ife_skbmark_ops = { + .metaid = IFE_META_SKBMARK, + .metatype = NLA_U32, + .name = "skbmark", + .synopsis = "skb mark 32 bit metadata", + .check_presence = skbmark_check, + .encode = skbmark_encode, + .decode = skbmark_decode, + .get = ife_get_meta_u32, + .alloc = ife_alloc_meta_u32, + .release = ife_release_meta_gen, + .validate = ife_validate_meta_u32, + .owner = THIS_MODULE, +}; + +static int __init ifemark_init_module(void) +{ + return register_ife_op(&ife_skbmark_ops); +} + +static void __exit ifemark_cleanup_module(void) +{ + unregister_ife_op(&ife_skbmark_ops); +} + +module_init(ifemark_init_module); +module_exit(ifemark_cleanup_module); + +MODULE_AUTHOR("Jamal Hadi Salim(2015)"); +MODULE_DESCRIPTION("Inter-FE skb mark metadata module"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_IFE_META("skbmark"); diff --git a/net/sched/act_meta_skbprio.c b/net/sched/act_meta_skbprio.c new file mode 100644 index 000000000..4033f9fc4 --- /dev/null +++ b/net/sched/act_meta_skbprio.c @@ -0,0 +1,76 @@ +/* + * net/sched/act_meta_prio.c IFE skb->priority metadata module + * + * 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. + * + * copyright Jamal Hadi Salim (2015) + * +*/ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <uapi/linux/tc_act/tc_ife.h> +#include <net/tc_act/tc_ife.h> + +static int skbprio_check(struct sk_buff *skb, struct tcf_meta_info *e) +{ + return ife_check_meta_u32(skb->priority, e); +} + +static int skbprio_encode(struct sk_buff *skb, void *skbdata, + struct tcf_meta_info *e) +{ + u32 ifeprio = skb->priority; /* avoid having to cast skb->priority*/ + + return ife_encode_meta_u32(ifeprio, skbdata, e); +} + +static int skbprio_decode(struct sk_buff *skb, void *data, u16 len) +{ + u32 ifeprio = *(u32 *)data; + + skb->priority = ntohl(ifeprio); + return 0; +} + +static struct tcf_meta_ops ife_prio_ops = { + .metaid = IFE_META_PRIO, + .metatype = NLA_U32, + .name = "skbprio", + .synopsis = "skb prio metadata", + .check_presence = skbprio_check, + .encode = skbprio_encode, + .decode = skbprio_decode, + .get = ife_get_meta_u32, + .alloc = ife_alloc_meta_u32, + .owner = THIS_MODULE, +}; + +static int __init ifeprio_init_module(void) +{ + return register_ife_op(&ife_prio_ops); +} + +static void __exit ifeprio_cleanup_module(void) +{ + unregister_ife_op(&ife_prio_ops); +} + +module_init(ifeprio_init_module); +module_exit(ifeprio_cleanup_module); + +MODULE_AUTHOR("Jamal Hadi Salim(2015)"); +MODULE_DESCRIPTION("Inter-FE skb prio metadata action"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_IFE_META("skbprio"); diff --git a/net/sched/act_meta_skbtcindex.c b/net/sched/act_meta_skbtcindex.c new file mode 100644 index 000000000..7221437ca --- /dev/null +++ b/net/sched/act_meta_skbtcindex.c @@ -0,0 +1,78 @@ +/* + * net/sched/act_meta_tc_index.c IFE skb->tc_index metadata module + * + * 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. + * + * copyright Jamal Hadi Salim (2016) + * +*/ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <uapi/linux/tc_act/tc_ife.h> +#include <net/tc_act/tc_ife.h> + +static int skbtcindex_encode(struct sk_buff *skb, void *skbdata, + struct tcf_meta_info *e) +{ + u32 ifetc_index = skb->tc_index; + + return ife_encode_meta_u16(ifetc_index, skbdata, e); +} + +static int skbtcindex_decode(struct sk_buff *skb, void *data, u16 len) +{ + u16 ifetc_index = *(u16 *)data; + + skb->tc_index = ntohs(ifetc_index); + return 0; +} + +static int skbtcindex_check(struct sk_buff *skb, struct tcf_meta_info *e) +{ + return ife_check_meta_u16(skb->tc_index, e); +} + +static struct tcf_meta_ops ife_skbtcindex_ops = { + .metaid = IFE_META_TCINDEX, + .metatype = NLA_U16, + .name = "tc_index", + .synopsis = "skb tc_index 16 bit metadata", + .check_presence = skbtcindex_check, + .encode = skbtcindex_encode, + .decode = skbtcindex_decode, + .get = ife_get_meta_u16, + .alloc = ife_alloc_meta_u16, + .release = ife_release_meta_gen, + .validate = ife_validate_meta_u16, + .owner = THIS_MODULE, +}; + +static int __init ifetc_index_init_module(void) +{ + return register_ife_op(&ife_skbtcindex_ops); +} + +static void __exit ifetc_index_cleanup_module(void) +{ + unregister_ife_op(&ife_skbtcindex_ops); +} + +module_init(ifetc_index_init_module); +module_exit(ifetc_index_cleanup_module); + +MODULE_AUTHOR("Jamal Hadi Salim(2016)"); +MODULE_DESCRIPTION("Inter-FE skb tc_index metadata module"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_IFE_META("tcindex"); diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c new file mode 100644 index 000000000..a30c17a28 --- /dev/null +++ b/net/sched/act_mirred.c @@ -0,0 +1,462 @@ +/* + * net/sched/act_mirred.c packet mirroring and redirect actions + * + * 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. + * + * Authors: Jamal Hadi Salim (2002-4) + * + * TODO: Add ingress support (and socket redirect support) + * + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/gfp.h> +#include <linux/if_arp.h> +#include <net/net_namespace.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <linux/tc_act/tc_mirred.h> +#include <net/tc_act/tc_mirred.h> + +static LIST_HEAD(mirred_list); +static DEFINE_SPINLOCK(mirred_list_lock); + +static bool tcf_mirred_is_act_redirect(int action) +{ + return action == TCA_EGRESS_REDIR || action == TCA_INGRESS_REDIR; +} + +static bool tcf_mirred_act_wants_ingress(int action) +{ + switch (action) { + case TCA_EGRESS_REDIR: + case TCA_EGRESS_MIRROR: + return false; + case TCA_INGRESS_REDIR: + case TCA_INGRESS_MIRROR: + return true; + default: + BUG(); + } +} + +static bool tcf_mirred_can_reinsert(int action) +{ + switch (action) { + case TC_ACT_SHOT: + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + return true; + } + return false; +} + +static struct net_device *tcf_mirred_dev_dereference(struct tcf_mirred *m) +{ + return rcu_dereference_protected(m->tcfm_dev, + lockdep_is_held(&m->tcf_lock)); +} + +static void tcf_mirred_release(struct tc_action *a) +{ + struct tcf_mirred *m = to_mirred(a); + struct net_device *dev; + + spin_lock(&mirred_list_lock); + list_del(&m->tcfm_list); + spin_unlock(&mirred_list_lock); + + /* last reference to action, no need to lock */ + dev = rcu_dereference_protected(m->tcfm_dev, 1); + if (dev) + dev_put(dev); +} + +static const struct nla_policy mirred_policy[TCA_MIRRED_MAX + 1] = { + [TCA_MIRRED_PARMS] = { .len = sizeof(struct tc_mirred) }, +}; + +static unsigned int mirred_net_id; +static struct tc_action_ops act_mirred_ops; + +static int tcf_mirred_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, mirred_net_id); + struct nlattr *tb[TCA_MIRRED_MAX + 1]; + bool mac_header_xmit = false; + struct tc_mirred *parm; + struct tcf_mirred *m; + struct net_device *dev; + bool exists = false; + int ret, err; + u32 index; + + if (!nla) { + NL_SET_ERR_MSG_MOD(extack, "Mirred requires attributes to be passed"); + return -EINVAL; + } + ret = nla_parse_nested(tb, TCA_MIRRED_MAX, nla, mirred_policy, extack); + if (ret < 0) + return ret; + if (!tb[TCA_MIRRED_PARMS]) { + NL_SET_ERR_MSG_MOD(extack, "Missing required mirred parameters"); + return -EINVAL; + } + parm = nla_data(tb[TCA_MIRRED_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + switch (parm->eaction) { + case TCA_EGRESS_MIRROR: + case TCA_EGRESS_REDIR: + case TCA_INGRESS_REDIR: + case TCA_INGRESS_MIRROR: + break; + default: + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + NL_SET_ERR_MSG_MOD(extack, "Unknown mirred option"); + return -EINVAL; + } + + if (!exists) { + if (!parm->ifindex) { + tcf_idr_cleanup(tn, index); + NL_SET_ERR_MSG_MOD(extack, "Specified device does not exist"); + return -EINVAL; + } + ret = tcf_idr_create(tn, index, est, a, + &act_mirred_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + m = to_mirred(*a); + + if (ret == ACT_P_CREATED) + INIT_LIST_HEAD(&m->tcfm_list); + + spin_lock_bh(&m->tcf_lock); + m->tcf_action = parm->action; + m->tcfm_eaction = parm->eaction; + + if (parm->ifindex) { + dev = dev_get_by_index(net, parm->ifindex); + if (!dev) { + spin_unlock_bh(&m->tcf_lock); + tcf_idr_release(*a, bind); + return -ENODEV; + } + mac_header_xmit = dev_is_mac_header_xmit(dev); + rcu_swap_protected(m->tcfm_dev, dev, + lockdep_is_held(&m->tcf_lock)); + if (dev) + dev_put(dev); + m->tcfm_mac_header_xmit = mac_header_xmit; + } + spin_unlock_bh(&m->tcf_lock); + + if (ret == ACT_P_CREATED) { + spin_lock(&mirred_list_lock); + list_add(&m->tcfm_list, &mirred_list); + spin_unlock(&mirred_list_lock); + + tcf_idr_insert(tn, *a); + } + + return ret; +} + +static int tcf_mirred_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_mirred *m = to_mirred(a); + struct sk_buff *skb2 = skb; + bool m_mac_header_xmit; + struct net_device *dev; + int retval, err = 0; + bool use_reinsert; + bool want_ingress; + bool is_redirect; + int m_eaction; + int mac_len; + + tcf_lastuse_update(&m->tcf_tm); + bstats_cpu_update(this_cpu_ptr(m->common.cpu_bstats), skb); + + m_mac_header_xmit = READ_ONCE(m->tcfm_mac_header_xmit); + m_eaction = READ_ONCE(m->tcfm_eaction); + retval = READ_ONCE(m->tcf_action); + dev = rcu_dereference_bh(m->tcfm_dev); + if (unlikely(!dev)) { + pr_notice_once("tc mirred: target device is gone\n"); + goto out; + } + + if (unlikely(!(dev->flags & IFF_UP))) { + net_notice_ratelimited("tc mirred to Houston: device %s is down\n", + dev->name); + goto out; + } + + /* we could easily avoid the clone only if called by ingress and clsact; + * since we can't easily detect the clsact caller, skip clone only for + * ingress - that covers the TC S/W datapath. + */ + is_redirect = tcf_mirred_is_act_redirect(m_eaction); + use_reinsert = skb_at_tc_ingress(skb) && is_redirect && + tcf_mirred_can_reinsert(retval); + if (!use_reinsert) { + skb2 = skb_clone(skb, GFP_ATOMIC); + if (!skb2) + goto out; + } + + /* If action's target direction differs than filter's direction, + * and devices expect a mac header on xmit, then mac push/pull is + * needed. + */ + want_ingress = tcf_mirred_act_wants_ingress(m_eaction); + if (skb_at_tc_ingress(skb) != want_ingress && m_mac_header_xmit) { + if (!skb_at_tc_ingress(skb)) { + /* caught at egress, act ingress: pull mac */ + mac_len = skb_network_header(skb) - skb_mac_header(skb); + skb_pull_rcsum(skb2, mac_len); + } else { + /* caught at ingress, act egress: push mac */ + skb_push_rcsum(skb2, skb->mac_len); + } + } + + skb2->skb_iif = skb->dev->ifindex; + skb2->dev = dev; + + /* mirror is always swallowed */ + if (is_redirect) { + skb2->tc_redirected = 1; + skb2->tc_from_ingress = skb2->tc_at_ingress; + + /* let's the caller reinsert the packet, if possible */ + if (use_reinsert) { + res->ingress = want_ingress; + res->qstats = this_cpu_ptr(m->common.cpu_qstats); + return TC_ACT_REINSERT; + } + } + + if (!want_ingress) + err = dev_queue_xmit(skb2); + else + err = netif_receive_skb(skb2); + + if (err) { +out: + qstats_overlimit_inc(this_cpu_ptr(m->common.cpu_qstats)); + if (tcf_mirred_is_act_redirect(m_eaction)) + retval = TC_ACT_SHOT; + } + + return retval; +} + +static void tcf_stats_update(struct tc_action *a, u64 bytes, u32 packets, + u64 lastuse) +{ + struct tcf_mirred *m = to_mirred(a); + struct tcf_t *tm = &m->tcf_tm; + + _bstats_cpu_update(this_cpu_ptr(a->cpu_bstats), bytes, packets); + tm->lastuse = max_t(u64, tm->lastuse, lastuse); +} + +static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_mirred *m = to_mirred(a); + struct tc_mirred opt = { + .index = m->tcf_index, + .refcnt = refcount_read(&m->tcf_refcnt) - ref, + .bindcnt = atomic_read(&m->tcf_bindcnt) - bind, + }; + struct net_device *dev; + struct tcf_t t; + + spin_lock_bh(&m->tcf_lock); + opt.action = m->tcf_action; + opt.eaction = m->tcfm_eaction; + dev = tcf_mirred_dev_dereference(m); + if (dev) + opt.ifindex = dev->ifindex; + + if (nla_put(skb, TCA_MIRRED_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + tcf_tm_dump(&t, &m->tcf_tm); + if (nla_put_64bit(skb, TCA_MIRRED_TM, sizeof(t), &t, TCA_MIRRED_PAD)) + goto nla_put_failure; + spin_unlock_bh(&m->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&m->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_mirred_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, mirred_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_mirred_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, mirred_net_id); + + return tcf_idr_search(tn, a, index); +} + +static int mirred_device_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct tcf_mirred *m; + + ASSERT_RTNL(); + if (event == NETDEV_UNREGISTER) { + spin_lock(&mirred_list_lock); + list_for_each_entry(m, &mirred_list, tcfm_list) { + spin_lock_bh(&m->tcf_lock); + if (tcf_mirred_dev_dereference(m) == dev) { + dev_put(dev); + /* Note : no rcu grace period necessary, as + * net_device are already rcu protected. + */ + RCU_INIT_POINTER(m->tcfm_dev, NULL); + } + spin_unlock_bh(&m->tcf_lock); + } + spin_unlock(&mirred_list_lock); + } + + return NOTIFY_DONE; +} + +static struct notifier_block mirred_device_notifier = { + .notifier_call = mirred_device_event, +}; + +static struct net_device *tcf_mirred_get_dev(const struct tc_action *a) +{ + struct tcf_mirred *m = to_mirred(a); + struct net_device *dev; + + rcu_read_lock(); + dev = rcu_dereference(m->tcfm_dev); + if (dev) + dev_hold(dev); + rcu_read_unlock(); + + return dev; +} + +static void tcf_mirred_put_dev(struct net_device *dev) +{ + dev_put(dev); +} + +static struct tc_action_ops act_mirred_ops = { + .kind = "mirred", + .type = TCA_ACT_MIRRED, + .owner = THIS_MODULE, + .act = tcf_mirred_act, + .stats_update = tcf_stats_update, + .dump = tcf_mirred_dump, + .cleanup = tcf_mirred_release, + .init = tcf_mirred_init, + .walk = tcf_mirred_walker, + .lookup = tcf_mirred_search, + .size = sizeof(struct tcf_mirred), + .get_dev = tcf_mirred_get_dev, + .put_dev = tcf_mirred_put_dev, +}; + +static __net_init int mirred_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, mirred_net_id); + + return tc_action_net_init(net, tn, &act_mirred_ops); +} + +static void __net_exit mirred_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, mirred_net_id); +} + +static struct pernet_operations mirred_net_ops = { + .init = mirred_init_net, + .exit_batch = mirred_exit_net, + .id = &mirred_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2002)"); +MODULE_DESCRIPTION("Device Mirror/redirect actions"); +MODULE_LICENSE("GPL"); + +static int __init mirred_init_module(void) +{ + int err = register_netdevice_notifier(&mirred_device_notifier); + if (err) + return err; + + pr_info("Mirror/redirect action on\n"); + err = tcf_register_action(&act_mirred_ops, &mirred_net_ops); + if (err) + unregister_netdevice_notifier(&mirred_device_notifier); + + return err; +} + +static void __exit mirred_cleanup_module(void) +{ + tcf_unregister_action(&act_mirred_ops, &mirred_net_ops); + unregister_netdevice_notifier(&mirred_device_notifier); +} + +module_init(mirred_init_module); +module_exit(mirred_cleanup_module); diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c new file mode 100644 index 000000000..d1b47a1b1 --- /dev/null +++ b/net/sched/act_nat.c @@ -0,0 +1,349 @@ +/* + * Stateless NAT actions + * + * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the Free + * Software Foundation; either version 2 of the License, or (at your option) + * any later version. + */ + +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/netfilter.h> +#include <linux/rtnetlink.h> +#include <linux/skbuff.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/string.h> +#include <linux/tc_act/tc_nat.h> +#include <net/act_api.h> +#include <net/icmp.h> +#include <net/ip.h> +#include <net/netlink.h> +#include <net/tc_act/tc_nat.h> +#include <net/tcp.h> +#include <net/udp.h> + + +static unsigned int nat_net_id; +static struct tc_action_ops act_nat_ops; + +static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = { + [TCA_NAT_PARMS] = { .len = sizeof(struct tc_nat) }, +}; + +static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, + struct tc_action **a, int ovr, int bind, + bool rtnl_held, struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, nat_net_id); + struct nlattr *tb[TCA_NAT_MAX + 1]; + struct tc_nat *parm; + int ret = 0, err; + struct tcf_nat *p; + u32 index; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_NAT_MAX, nla, nat_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_NAT_PARMS] == NULL) + return -EINVAL; + parm = nla_data(tb[TCA_NAT_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (!err) { + ret = tcf_idr_create(tn, index, est, a, + &act_nat_ops, bind, false); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else if (err > 0) { + if (bind) + return 0; + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + } else { + return err; + } + p = to_tcf_nat(*a); + + spin_lock_bh(&p->tcf_lock); + p->old_addr = parm->old_addr; + p->new_addr = parm->new_addr; + p->mask = parm->mask; + p->flags = parm->flags; + + p->tcf_action = parm->action; + spin_unlock_bh(&p->tcf_lock); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + + return ret; +} + +static int tcf_nat_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_nat *p = to_tcf_nat(a); + struct iphdr *iph; + __be32 old_addr; + __be32 new_addr; + __be32 mask; + __be32 addr; + int egress; + int action; + int ihl; + int noff; + + spin_lock(&p->tcf_lock); + + tcf_lastuse_update(&p->tcf_tm); + old_addr = p->old_addr; + new_addr = p->new_addr; + mask = p->mask; + egress = p->flags & TCA_NAT_FLAG_EGRESS; + action = p->tcf_action; + + bstats_update(&p->tcf_bstats, skb); + + spin_unlock(&p->tcf_lock); + + if (unlikely(action == TC_ACT_SHOT)) + goto drop; + + noff = skb_network_offset(skb); + if (!pskb_may_pull(skb, sizeof(*iph) + noff)) + goto drop; + + iph = ip_hdr(skb); + + if (egress) + addr = iph->saddr; + else + addr = iph->daddr; + + if (!((old_addr ^ addr) & mask)) { + if (skb_try_make_writable(skb, sizeof(*iph) + noff)) + goto drop; + + new_addr &= mask; + new_addr |= addr & ~mask; + + /* Rewrite IP header */ + iph = ip_hdr(skb); + if (egress) + iph->saddr = new_addr; + else + iph->daddr = new_addr; + + csum_replace4(&iph->check, addr, new_addr); + } else if ((iph->frag_off & htons(IP_OFFSET)) || + iph->protocol != IPPROTO_ICMP) { + goto out; + } + + ihl = iph->ihl * 4; + + /* It would be nice to share code with stateful NAT. */ + switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) { + case IPPROTO_TCP: + { + struct tcphdr *tcph; + + if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) || + skb_try_make_writable(skb, ihl + sizeof(*tcph) + noff)) + goto drop; + + tcph = (void *)(skb_network_header(skb) + ihl); + inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr, + true); + break; + } + case IPPROTO_UDP: + { + struct udphdr *udph; + + if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) || + skb_try_make_writable(skb, ihl + sizeof(*udph) + noff)) + goto drop; + + udph = (void *)(skb_network_header(skb) + ihl); + if (udph->check || skb->ip_summed == CHECKSUM_PARTIAL) { + inet_proto_csum_replace4(&udph->check, skb, addr, + new_addr, true); + if (!udph->check) + udph->check = CSUM_MANGLED_0; + } + break; + } + case IPPROTO_ICMP: + { + struct icmphdr *icmph; + + if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + noff)) + goto drop; + + icmph = (void *)(skb_network_header(skb) + ihl); + + if ((icmph->type != ICMP_DEST_UNREACH) && + (icmph->type != ICMP_TIME_EXCEEDED) && + (icmph->type != ICMP_PARAMETERPROB)) + break; + + if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + sizeof(*iph) + + noff)) + goto drop; + + icmph = (void *)(skb_network_header(skb) + ihl); + iph = (void *)(icmph + 1); + if (egress) + addr = iph->daddr; + else + addr = iph->saddr; + + if ((old_addr ^ addr) & mask) + break; + + if (skb_try_make_writable(skb, ihl + sizeof(*icmph) + + sizeof(*iph) + noff)) + goto drop; + + icmph = (void *)(skb_network_header(skb) + ihl); + iph = (void *)(icmph + 1); + + new_addr &= mask; + new_addr |= addr & ~mask; + + /* XXX Fix up the inner checksums. */ + if (egress) + iph->daddr = new_addr; + else + iph->saddr = new_addr; + + inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr, + false); + break; + } + default: + break; + } + +out: + return action; + +drop: + spin_lock(&p->tcf_lock); + p->tcf_qstats.drops++; + spin_unlock(&p->tcf_lock); + return TC_ACT_SHOT; +} + +static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_nat *p = to_tcf_nat(a); + struct tc_nat opt = { + .old_addr = p->old_addr, + .new_addr = p->new_addr, + .mask = p->mask, + .flags = p->flags, + + .index = p->tcf_index, + .action = p->tcf_action, + .refcnt = refcount_read(&p->tcf_refcnt) - ref, + .bindcnt = atomic_read(&p->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + if (nla_put(skb, TCA_NAT_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + tcf_tm_dump(&t, &p->tcf_tm); + if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD)) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_nat_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, nat_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_nat_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, nat_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_nat_ops = { + .kind = "nat", + .type = TCA_ACT_NAT, + .owner = THIS_MODULE, + .act = tcf_nat_act, + .dump = tcf_nat_dump, + .init = tcf_nat_init, + .walk = tcf_nat_walker, + .lookup = tcf_nat_search, + .size = sizeof(struct tcf_nat), +}; + +static __net_init int nat_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, nat_net_id); + + return tc_action_net_init(net, tn, &act_nat_ops); +} + +static void __net_exit nat_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, nat_net_id); +} + +static struct pernet_operations nat_net_ops = { + .init = nat_init_net, + .exit_batch = nat_exit_net, + .id = &nat_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_DESCRIPTION("Stateless NAT actions"); +MODULE_LICENSE("GPL"); + +static int __init nat_init_module(void) +{ + return tcf_register_action(&act_nat_ops, &nat_net_ops); +} + +static void __exit nat_cleanup_module(void) +{ + tcf_unregister_action(&act_nat_ops, &nat_net_ops); +} + +module_init(nat_init_module); +module_exit(nat_cleanup_module); diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c new file mode 100644 index 000000000..aeb8f84cb --- /dev/null +++ b/net/sched/act_pedit.c @@ -0,0 +1,541 @@ +/* + * net/sched/act_pedit.c Generic packet editor + * + * 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. + * + * Authors: Jamal Hadi Salim (2002-4) + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <linux/tc_act/tc_pedit.h> +#include <net/tc_act/tc_pedit.h> +#include <uapi/linux/tc_act/tc_pedit.h> + +static unsigned int pedit_net_id; +static struct tc_action_ops act_pedit_ops; + +static const struct nla_policy pedit_policy[TCA_PEDIT_MAX + 1] = { + [TCA_PEDIT_PARMS] = { .len = sizeof(struct tc_pedit) }, + [TCA_PEDIT_KEYS_EX] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy pedit_key_ex_policy[TCA_PEDIT_KEY_EX_MAX + 1] = { + [TCA_PEDIT_KEY_EX_HTYPE] = { .type = NLA_U16 }, + [TCA_PEDIT_KEY_EX_CMD] = { .type = NLA_U16 }, +}; + +static struct tcf_pedit_key_ex *tcf_pedit_keys_ex_parse(struct nlattr *nla, + u8 n) +{ + struct tcf_pedit_key_ex *keys_ex; + struct tcf_pedit_key_ex *k; + const struct nlattr *ka; + int err = -EINVAL; + int rem; + + if (!nla) + return NULL; + + keys_ex = kcalloc(n, sizeof(*k), GFP_KERNEL); + if (!keys_ex) + return ERR_PTR(-ENOMEM); + + k = keys_ex; + + nla_for_each_nested(ka, nla, rem) { + struct nlattr *tb[TCA_PEDIT_KEY_EX_MAX + 1]; + + if (!n) { + err = -EINVAL; + goto err_out; + } + n--; + + if (nla_type(ka) != TCA_PEDIT_KEY_EX) { + err = -EINVAL; + goto err_out; + } + + err = nla_parse_nested(tb, TCA_PEDIT_KEY_EX_MAX, ka, + pedit_key_ex_policy, NULL); + if (err) + goto err_out; + + if (!tb[TCA_PEDIT_KEY_EX_HTYPE] || + !tb[TCA_PEDIT_KEY_EX_CMD]) { + err = -EINVAL; + goto err_out; + } + + k->htype = nla_get_u16(tb[TCA_PEDIT_KEY_EX_HTYPE]); + k->cmd = nla_get_u16(tb[TCA_PEDIT_KEY_EX_CMD]); + + if (k->htype > TCA_PEDIT_HDR_TYPE_MAX || + k->cmd > TCA_PEDIT_CMD_MAX) { + err = -EINVAL; + goto err_out; + } + + k++; + } + + if (n) { + err = -EINVAL; + goto err_out; + } + + return keys_ex; + +err_out: + kfree(keys_ex); + return ERR_PTR(err); +} + +static int tcf_pedit_key_ex_dump(struct sk_buff *skb, + struct tcf_pedit_key_ex *keys_ex, int n) +{ + struct nlattr *keys_start = nla_nest_start(skb, TCA_PEDIT_KEYS_EX); + + if (!keys_start) + goto nla_failure; + for (; n > 0; n--) { + struct nlattr *key_start; + + key_start = nla_nest_start(skb, TCA_PEDIT_KEY_EX); + if (!key_start) + goto nla_failure; + + if (nla_put_u16(skb, TCA_PEDIT_KEY_EX_HTYPE, keys_ex->htype) || + nla_put_u16(skb, TCA_PEDIT_KEY_EX_CMD, keys_ex->cmd)) + goto nla_failure; + + nla_nest_end(skb, key_start); + + keys_ex++; + } + + nla_nest_end(skb, keys_start); + + return 0; +nla_failure: + nla_nest_cancel(skb, keys_start); + return -EINVAL; +} + +static int tcf_pedit_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, pedit_net_id); + struct nlattr *tb[TCA_PEDIT_MAX + 1]; + struct tc_pedit_key *keys = NULL; + struct tcf_pedit_key_ex *keys_ex; + struct tc_pedit *parm; + struct nlattr *pattr; + struct tcf_pedit *p; + int ret = 0, err; + int i, ksize; + u32 index; + + if (!nla) { + NL_SET_ERR_MSG_MOD(extack, "Pedit requires attributes to be passed"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_PEDIT_MAX, nla, pedit_policy, NULL); + if (err < 0) + return err; + + pattr = tb[TCA_PEDIT_PARMS]; + if (!pattr) + pattr = tb[TCA_PEDIT_PARMS_EX]; + if (!pattr) { + NL_SET_ERR_MSG_MOD(extack, "Missing required TCA_PEDIT_PARMS or TCA_PEDIT_PARMS_EX pedit attribute"); + return -EINVAL; + } + + parm = nla_data(pattr); + if (!parm->nkeys) { + NL_SET_ERR_MSG_MOD(extack, "Pedit requires keys to be passed"); + return -EINVAL; + } + ksize = parm->nkeys * sizeof(struct tc_pedit_key); + if (nla_len(pattr) < sizeof(*parm) + ksize) { + NL_SET_ERR_MSG_ATTR(extack, pattr, "Length of TCA_PEDIT_PARMS or TCA_PEDIT_PARMS_EX pedit attribute is invalid"); + return -EINVAL; + } + + keys_ex = tcf_pedit_keys_ex_parse(tb[TCA_PEDIT_KEYS_EX], parm->nkeys); + if (IS_ERR(keys_ex)) + return PTR_ERR(keys_ex); + + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (!err) { + ret = tcf_idr_create(tn, index, est, a, + &act_pedit_ops, bind, false); + if (ret) { + tcf_idr_cleanup(tn, index); + goto out_free; + } + ret = ACT_P_CREATED; + } else if (err > 0) { + if (bind) + goto out_free; + if (!ovr) { + ret = -EEXIST; + goto out_release; + } + } else { + ret = err; + goto out_free; + } + + p = to_pedit(*a); + spin_lock_bh(&p->tcf_lock); + + if (ret == ACT_P_CREATED || + (p->tcfp_nkeys && p->tcfp_nkeys != parm->nkeys)) { + keys = kmalloc(ksize, GFP_ATOMIC); + if (!keys) { + spin_unlock_bh(&p->tcf_lock); + ret = -ENOMEM; + goto out_release; + } + kfree(p->tcfp_keys); + p->tcfp_keys = keys; + p->tcfp_nkeys = parm->nkeys; + } + memcpy(p->tcfp_keys, parm->keys, ksize); + p->tcfp_off_max_hint = 0; + for (i = 0; i < p->tcfp_nkeys; ++i) { + u32 cur = p->tcfp_keys[i].off; + + /* sanitize the shift value for any later use */ + p->tcfp_keys[i].shift = min_t(size_t, BITS_PER_TYPE(int) - 1, + p->tcfp_keys[i].shift); + + /* The AT option can read a single byte, we can bound the actual + * value with uchar max. + */ + cur += (0xff & p->tcfp_keys[i].offmask) >> p->tcfp_keys[i].shift; + + /* Each key touches 4 bytes starting from the computed offset */ + p->tcfp_off_max_hint = max(p->tcfp_off_max_hint, cur + 4); + } + + p->tcfp_flags = parm->flags; + p->tcf_action = parm->action; + + kfree(p->tcfp_keys_ex); + p->tcfp_keys_ex = keys_ex; + + spin_unlock_bh(&p->tcf_lock); + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; + +out_release: + tcf_idr_release(*a, bind); +out_free: + kfree(keys_ex); + return ret; + +} + +static void tcf_pedit_cleanup(struct tc_action *a) +{ + struct tcf_pedit *p = to_pedit(a); + struct tc_pedit_key *keys = p->tcfp_keys; + + kfree(keys); + kfree(p->tcfp_keys_ex); +} + +static bool offset_valid(struct sk_buff *skb, int offset) +{ + if (offset > 0 && offset > skb->len) + return false; + + if (offset < 0 && -offset > skb_headroom(skb)) + return false; + + return true; +} + +static int pedit_skb_hdr_offset(struct sk_buff *skb, + enum pedit_header_type htype, int *hoffset) +{ + int ret = -EINVAL; + + switch (htype) { + case TCA_PEDIT_KEY_EX_HDR_TYPE_ETH: + if (skb_mac_header_was_set(skb)) { + *hoffset = skb_mac_offset(skb); + ret = 0; + } + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_NETWORK: + case TCA_PEDIT_KEY_EX_HDR_TYPE_IP4: + case TCA_PEDIT_KEY_EX_HDR_TYPE_IP6: + *hoffset = skb_network_offset(skb); + ret = 0; + break; + case TCA_PEDIT_KEY_EX_HDR_TYPE_TCP: + case TCA_PEDIT_KEY_EX_HDR_TYPE_UDP: + if (skb_transport_header_was_set(skb)) { + *hoffset = skb_transport_offset(skb); + ret = 0; + } + break; + default: + ret = -EINVAL; + break; + } + + return ret; +} + +static int tcf_pedit_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_pedit *p = to_pedit(a); + u32 max_offset; + int i; + + spin_lock(&p->tcf_lock); + + max_offset = (skb_transport_header_was_set(skb) ? + skb_transport_offset(skb) : + skb_network_offset(skb)) + + p->tcfp_off_max_hint; + if (skb_ensure_writable(skb, min(skb->len, max_offset))) + goto unlock; + + tcf_lastuse_update(&p->tcf_tm); + + if (p->tcfp_nkeys > 0) { + struct tc_pedit_key *tkey = p->tcfp_keys; + struct tcf_pedit_key_ex *tkey_ex = p->tcfp_keys_ex; + enum pedit_header_type htype = + TCA_PEDIT_KEY_EX_HDR_TYPE_NETWORK; + enum pedit_cmd cmd = TCA_PEDIT_KEY_EX_CMD_SET; + + for (i = p->tcfp_nkeys; i > 0; i--, tkey++) { + u32 *ptr, hdata; + int offset = tkey->off; + int hoffset; + u32 val; + int rc; + + if (tkey_ex) { + htype = tkey_ex->htype; + cmd = tkey_ex->cmd; + + tkey_ex++; + } + + rc = pedit_skb_hdr_offset(skb, htype, &hoffset); + if (rc) { + pr_info("tc action pedit bad header type specified (0x%x)\n", + htype); + goto bad; + } + + if (tkey->offmask) { + u8 *d, _d; + + if (!offset_valid(skb, hoffset + tkey->at)) { + pr_info("tc action pedit 'at' offset %d out of bounds\n", + hoffset + tkey->at); + goto bad; + } + d = skb_header_pointer(skb, hoffset + tkey->at, + sizeof(_d), &_d); + if (!d) + goto bad; + offset += (*d & tkey->offmask) >> tkey->shift; + } + + if (offset % 4) { + pr_info("tc action pedit offset must be on 32 bit boundaries\n"); + goto bad; + } + + if (!offset_valid(skb, hoffset + offset)) { + pr_info("tc action pedit offset %d out of bounds\n", + hoffset + offset); + goto bad; + } + + ptr = skb_header_pointer(skb, hoffset + offset, + sizeof(hdata), &hdata); + if (!ptr) + goto bad; + /* just do it, baby */ + switch (cmd) { + case TCA_PEDIT_KEY_EX_CMD_SET: + val = tkey->val; + break; + case TCA_PEDIT_KEY_EX_CMD_ADD: + val = (*ptr + tkey->val) & ~tkey->mask; + break; + default: + pr_info("tc action pedit bad command (%d)\n", + cmd); + goto bad; + } + + *ptr = ((*ptr & tkey->mask) ^ val); + if (ptr == &hdata) + skb_store_bits(skb, hoffset + offset, ptr, 4); + } + + goto done; + } else { + WARN(1, "pedit BUG: index %d\n", p->tcf_index); + } + +bad: + p->tcf_qstats.overlimits++; +done: + bstats_update(&p->tcf_bstats, skb); +unlock: + spin_unlock(&p->tcf_lock); + return p->tcf_action; +} + +static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_pedit *p = to_pedit(a); + struct tc_pedit *opt; + struct tcf_t t; + int s; + + s = sizeof(*opt) + p->tcfp_nkeys * sizeof(struct tc_pedit_key); + + /* netlink spinlocks held above us - must use ATOMIC */ + opt = kzalloc(s, GFP_ATOMIC); + if (unlikely(!opt)) + return -ENOBUFS; + + spin_lock_bh(&p->tcf_lock); + memcpy(opt->keys, p->tcfp_keys, + p->tcfp_nkeys * sizeof(struct tc_pedit_key)); + opt->index = p->tcf_index; + opt->nkeys = p->tcfp_nkeys; + opt->flags = p->tcfp_flags; + opt->action = p->tcf_action; + opt->refcnt = refcount_read(&p->tcf_refcnt) - ref; + opt->bindcnt = atomic_read(&p->tcf_bindcnt) - bind; + + if (p->tcfp_keys_ex) { + if (tcf_pedit_key_ex_dump(skb, + p->tcfp_keys_ex, + p->tcfp_nkeys)) + goto nla_put_failure; + + if (nla_put(skb, TCA_PEDIT_PARMS_EX, s, opt)) + goto nla_put_failure; + } else { + if (nla_put(skb, TCA_PEDIT_PARMS, s, opt)) + goto nla_put_failure; + } + + tcf_tm_dump(&t, &p->tcf_tm); + if (nla_put_64bit(skb, TCA_PEDIT_TM, sizeof(t), &t, TCA_PEDIT_PAD)) + goto nla_put_failure; + spin_unlock_bh(&p->tcf_lock); + + kfree(opt); + return skb->len; + +nla_put_failure: + spin_unlock_bh(&p->tcf_lock); + nlmsg_trim(skb, b); + kfree(opt); + return -1; +} + +static int tcf_pedit_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, pedit_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_pedit_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, pedit_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_pedit_ops = { + .kind = "pedit", + .type = TCA_ACT_PEDIT, + .owner = THIS_MODULE, + .act = tcf_pedit_act, + .dump = tcf_pedit_dump, + .cleanup = tcf_pedit_cleanup, + .init = tcf_pedit_init, + .walk = tcf_pedit_walker, + .lookup = tcf_pedit_search, + .size = sizeof(struct tcf_pedit), +}; + +static __net_init int pedit_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, pedit_net_id); + + return tc_action_net_init(net, tn, &act_pedit_ops); +} + +static void __net_exit pedit_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, pedit_net_id); +} + +static struct pernet_operations pedit_net_ops = { + .init = pedit_init_net, + .exit_batch = pedit_exit_net, + .id = &pedit_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2002-4)"); +MODULE_DESCRIPTION("Generic Packet Editor actions"); +MODULE_LICENSE("GPL"); + +static int __init pedit_init_module(void) +{ + return tcf_register_action(&act_pedit_ops, &pedit_net_ops); +} + +static void __exit pedit_cleanup_module(void) +{ + tcf_unregister_action(&act_pedit_ops, &pedit_net_ops); +} + +module_init(pedit_init_module); +module_exit(pedit_cleanup_module); diff --git a/net/sched/act_police.c b/net/sched/act_police.c new file mode 100644 index 000000000..4db25959e --- /dev/null +++ b/net/sched/act_police.c @@ -0,0 +1,371 @@ +/* + * net/sched/act_police.c Input police filter + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * J Hadi Salim (action changes) + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <net/act_api.h> +#include <net/netlink.h> + +struct tcf_police { + struct tc_action common; + int tcfp_result; + u32 tcfp_ewma_rate; + s64 tcfp_burst; + u32 tcfp_mtu; + s64 tcfp_toks; + s64 tcfp_ptoks; + s64 tcfp_mtu_ptoks; + s64 tcfp_t_c; + struct psched_ratecfg rate; + bool rate_present; + struct psched_ratecfg peak; + bool peak_present; +}; + +#define to_police(pc) ((struct tcf_police *)pc) + +/* old policer structure from before tc actions */ +struct tc_police_compat { + u32 index; + int action; + u32 limit; + u32 burst; + u32 mtu; + struct tc_ratespec rate; + struct tc_ratespec peakrate; +}; + +/* Each policer is serialized by its individual spinlock */ + +static unsigned int police_net_id; +static struct tc_action_ops act_police_ops; + +static int tcf_police_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, police_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static const struct nla_policy police_policy[TCA_POLICE_MAX + 1] = { + [TCA_POLICE_RATE] = { .len = TC_RTAB_SIZE }, + [TCA_POLICE_PEAKRATE] = { .len = TC_RTAB_SIZE }, + [TCA_POLICE_AVRATE] = { .type = NLA_U32 }, + [TCA_POLICE_RESULT] = { .type = NLA_U32 }, +}; + +static int tcf_police_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + int ret = 0, err; + struct nlattr *tb[TCA_POLICE_MAX + 1]; + struct tc_police *parm; + struct tcf_police *police; + struct qdisc_rate_table *R_tab = NULL, *P_tab = NULL; + struct tc_action_net *tn = net_generic(net, police_net_id); + bool exists = false; + u32 index; + int size; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_POLICE_MAX, nla, police_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_POLICE_TBF] == NULL) + return -EINVAL; + size = nla_len(tb[TCA_POLICE_TBF]); + if (size != sizeof(*parm) && size != sizeof(struct tc_police_compat)) + return -EINVAL; + + parm = nla_data(tb[TCA_POLICE_TBF]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + if (!exists) { + ret = tcf_idr_create(tn, index, NULL, a, + &act_police_ops, bind, false); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + + police = to_police(*a); + if (parm->rate.rate) { + err = -ENOMEM; + R_tab = qdisc_get_rtab(&parm->rate, tb[TCA_POLICE_RATE], NULL); + if (R_tab == NULL) + goto failure; + + if (parm->peakrate.rate) { + P_tab = qdisc_get_rtab(&parm->peakrate, + tb[TCA_POLICE_PEAKRATE], NULL); + if (P_tab == NULL) + goto failure; + } + } + + if (est) { + err = gen_replace_estimator(&police->tcf_bstats, NULL, + &police->tcf_rate_est, + &police->tcf_lock, + NULL, est); + if (err) + goto failure; + } else if (tb[TCA_POLICE_AVRATE] && + (ret == ACT_P_CREATED || + !gen_estimator_active(&police->tcf_rate_est))) { + err = -EINVAL; + goto failure; + } + + spin_lock_bh(&police->tcf_lock); + /* No failure allowed after this point */ + police->tcfp_mtu = parm->mtu; + if (police->tcfp_mtu == 0) { + police->tcfp_mtu = ~0; + if (R_tab) + police->tcfp_mtu = 255 << R_tab->rate.cell_log; + } + if (R_tab) { + police->rate_present = true; + psched_ratecfg_precompute(&police->rate, &R_tab->rate, 0); + qdisc_put_rtab(R_tab); + } else { + police->rate_present = false; + } + if (P_tab) { + police->peak_present = true; + psched_ratecfg_precompute(&police->peak, &P_tab->rate, 0); + qdisc_put_rtab(P_tab); + } else { + police->peak_present = false; + } + + if (tb[TCA_POLICE_RESULT]) + police->tcfp_result = nla_get_u32(tb[TCA_POLICE_RESULT]); + police->tcfp_burst = PSCHED_TICKS2NS(parm->burst); + police->tcfp_toks = police->tcfp_burst; + if (police->peak_present) { + police->tcfp_mtu_ptoks = (s64) psched_l2t_ns(&police->peak, + police->tcfp_mtu); + police->tcfp_ptoks = police->tcfp_mtu_ptoks; + } + police->tcf_action = parm->action; + + if (tb[TCA_POLICE_AVRATE]) + police->tcfp_ewma_rate = nla_get_u32(tb[TCA_POLICE_AVRATE]); + + spin_unlock_bh(&police->tcf_lock); + if (ret != ACT_P_CREATED) + return ret; + + police->tcfp_t_c = ktime_get_ns(); + tcf_idr_insert(tn, *a); + + return ret; + +failure: + qdisc_put_rtab(P_tab); + qdisc_put_rtab(R_tab); + tcf_idr_release(*a, bind); + return err; +} + +static int tcf_police_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_police *police = to_police(a); + s64 now; + s64 toks; + s64 ptoks = 0; + + spin_lock(&police->tcf_lock); + + bstats_update(&police->tcf_bstats, skb); + tcf_lastuse_update(&police->tcf_tm); + + if (police->tcfp_ewma_rate) { + struct gnet_stats_rate_est64 sample; + + if (!gen_estimator_read(&police->tcf_rate_est, &sample) || + sample.bps >= police->tcfp_ewma_rate) { + police->tcf_qstats.overlimits++; + if (police->tcf_action == TC_ACT_SHOT) + police->tcf_qstats.drops++; + spin_unlock(&police->tcf_lock); + return police->tcf_action; + } + } + + if (qdisc_pkt_len(skb) <= police->tcfp_mtu) { + if (!police->rate_present) { + spin_unlock(&police->tcf_lock); + return police->tcfp_result; + } + + now = ktime_get_ns(); + toks = min_t(s64, now - police->tcfp_t_c, + police->tcfp_burst); + if (police->peak_present) { + ptoks = toks + police->tcfp_ptoks; + if (ptoks > police->tcfp_mtu_ptoks) + ptoks = police->tcfp_mtu_ptoks; + ptoks -= (s64) psched_l2t_ns(&police->peak, + qdisc_pkt_len(skb)); + } + toks += police->tcfp_toks; + if (toks > police->tcfp_burst) + toks = police->tcfp_burst; + toks -= (s64) psched_l2t_ns(&police->rate, qdisc_pkt_len(skb)); + if ((toks|ptoks) >= 0) { + police->tcfp_t_c = now; + police->tcfp_toks = toks; + police->tcfp_ptoks = ptoks; + if (police->tcfp_result == TC_ACT_SHOT) + police->tcf_qstats.drops++; + spin_unlock(&police->tcf_lock); + return police->tcfp_result; + } + } + + police->tcf_qstats.overlimits++; + if (police->tcf_action == TC_ACT_SHOT) + police->tcf_qstats.drops++; + spin_unlock(&police->tcf_lock); + return police->tcf_action; +} + +static int tcf_police_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_police *police = to_police(a); + struct tc_police opt = { + .index = police->tcf_index, + .refcnt = refcount_read(&police->tcf_refcnt) - ref, + .bindcnt = atomic_read(&police->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&police->tcf_lock); + opt.action = police->tcf_action; + opt.mtu = police->tcfp_mtu; + opt.burst = PSCHED_NS2TICKS(police->tcfp_burst); + if (police->rate_present) + psched_ratecfg_getrate(&opt.rate, &police->rate); + if (police->peak_present) + psched_ratecfg_getrate(&opt.peakrate, &police->peak); + if (nla_put(skb, TCA_POLICE_TBF, sizeof(opt), &opt)) + goto nla_put_failure; + if (police->tcfp_result && + nla_put_u32(skb, TCA_POLICE_RESULT, police->tcfp_result)) + goto nla_put_failure; + if (police->tcfp_ewma_rate && + nla_put_u32(skb, TCA_POLICE_AVRATE, police->tcfp_ewma_rate)) + goto nla_put_failure; + + t.install = jiffies_to_clock_t(jiffies - police->tcf_tm.install); + t.lastuse = jiffies_to_clock_t(jiffies - police->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - police->tcf_tm.firstuse); + t.expires = jiffies_to_clock_t(police->tcf_tm.expires); + if (nla_put_64bit(skb, TCA_POLICE_TM, sizeof(t), &t, TCA_POLICE_PAD)) + goto nla_put_failure; + spin_unlock_bh(&police->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&police->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_police_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, police_net_id); + + return tcf_idr_search(tn, a, index); +} + +MODULE_AUTHOR("Alexey Kuznetsov"); +MODULE_DESCRIPTION("Policing actions"); +MODULE_LICENSE("GPL"); + +static struct tc_action_ops act_police_ops = { + .kind = "police", + .type = TCA_ID_POLICE, + .owner = THIS_MODULE, + .act = tcf_police_act, + .dump = tcf_police_dump, + .init = tcf_police_init, + .walk = tcf_police_walker, + .lookup = tcf_police_search, + .size = sizeof(struct tcf_police), +}; + +static __net_init int police_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, police_net_id); + + return tc_action_net_init(net, tn, &act_police_ops); +} + +static void __net_exit police_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, police_net_id); +} + +static struct pernet_operations police_net_ops = { + .init = police_init_net, + .exit_batch = police_exit_net, + .id = &police_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init police_init_module(void) +{ + return tcf_register_action(&act_police_ops, &police_net_ops); +} + +static void __exit police_cleanup_module(void) +{ + tcf_unregister_action(&act_police_ops, &police_net_ops); +} + +module_init(police_init_module); +module_exit(police_cleanup_module); diff --git a/net/sched/act_sample.c b/net/sched/act_sample.c new file mode 100644 index 000000000..ea0738ceb --- /dev/null +++ b/net/sched/act_sample.c @@ -0,0 +1,292 @@ +/* + * net/sched/act_sample.c - Packet sampling tc action + * Copyright (c) 2017 Yotam Gigi <yotamg@mellanox.com> + * + * 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/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/gfp.h> +#include <net/net_namespace.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <linux/tc_act/tc_sample.h> +#include <net/tc_act/tc_sample.h> +#include <net/psample.h> + +#include <linux/if_arp.h> + +static unsigned int sample_net_id; +static struct tc_action_ops act_sample_ops; + +static const struct nla_policy sample_policy[TCA_SAMPLE_MAX + 1] = { + [TCA_SAMPLE_PARMS] = { .len = sizeof(struct tc_sample) }, + [TCA_SAMPLE_RATE] = { .type = NLA_U32 }, + [TCA_SAMPLE_TRUNC_SIZE] = { .type = NLA_U32 }, + [TCA_SAMPLE_PSAMPLE_GROUP] = { .type = NLA_U32 }, +}; + +static int tcf_sample_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, int ovr, + int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, sample_net_id); + struct nlattr *tb[TCA_SAMPLE_MAX + 1]; + struct psample_group *psample_group; + u32 psample_group_num, rate, index; + struct tc_sample *parm; + struct tcf_sample *s; + bool exists = false; + int ret, err; + + if (!nla) + return -EINVAL; + ret = nla_parse_nested(tb, TCA_SAMPLE_MAX, nla, sample_policy, NULL); + if (ret < 0) + return ret; + if (!tb[TCA_SAMPLE_PARMS] || !tb[TCA_SAMPLE_RATE] || + !tb[TCA_SAMPLE_PSAMPLE_GROUP]) + return -EINVAL; + + parm = nla_data(tb[TCA_SAMPLE_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, + &act_sample_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + ret = ACT_P_CREATED; + } else if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + + rate = nla_get_u32(tb[TCA_SAMPLE_RATE]); + if (!rate) { + NL_SET_ERR_MSG(extack, "invalid sample rate"); + tcf_idr_release(*a, bind); + return -EINVAL; + } + psample_group_num = nla_get_u32(tb[TCA_SAMPLE_PSAMPLE_GROUP]); + psample_group = psample_group_get(net, psample_group_num); + if (!psample_group) { + tcf_idr_release(*a, bind); + return -ENOMEM; + } + + s = to_sample(*a); + + spin_lock_bh(&s->tcf_lock); + s->tcf_action = parm->action; + s->rate = rate; + s->psample_group_num = psample_group_num; + rcu_swap_protected(s->psample_group, psample_group, + lockdep_is_held(&s->tcf_lock)); + + if (tb[TCA_SAMPLE_TRUNC_SIZE]) { + s->truncate = true; + s->trunc_size = nla_get_u32(tb[TCA_SAMPLE_TRUNC_SIZE]); + } + spin_unlock_bh(&s->tcf_lock); + + if (psample_group) + psample_group_put(psample_group); + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; +} + +static void tcf_sample_cleanup(struct tc_action *a) +{ + struct tcf_sample *s = to_sample(a); + struct psample_group *psample_group; + + /* last reference to action, no need to lock */ + psample_group = rcu_dereference_protected(s->psample_group, 1); + RCU_INIT_POINTER(s->psample_group, NULL); + if (psample_group) + psample_group_put(psample_group); +} + +static bool tcf_sample_dev_ok_push(struct net_device *dev) +{ + switch (dev->type) { + case ARPHRD_TUNNEL: + case ARPHRD_TUNNEL6: + case ARPHRD_SIT: + case ARPHRD_IPGRE: + case ARPHRD_IP6GRE: + case ARPHRD_VOID: + case ARPHRD_NONE: + return false; + default: + return true; + } +} + +static int tcf_sample_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_sample *s = to_sample(a); + struct psample_group *psample_group; + int retval; + int size; + int iif; + int oif; + + tcf_lastuse_update(&s->tcf_tm); + bstats_cpu_update(this_cpu_ptr(s->common.cpu_bstats), skb); + retval = READ_ONCE(s->tcf_action); + + psample_group = rcu_dereference_bh(s->psample_group); + + /* randomly sample packets according to rate */ + if (psample_group && (prandom_u32() % s->rate == 0)) { + if (!skb_at_tc_ingress(skb)) { + iif = skb->skb_iif; + oif = skb->dev->ifindex; + } else { + iif = skb->dev->ifindex; + oif = 0; + } + + /* on ingress, the mac header gets popped, so push it back */ + if (skb_at_tc_ingress(skb) && tcf_sample_dev_ok_push(skb->dev)) + skb_push(skb, skb->mac_len); + + size = s->truncate ? s->trunc_size : skb->len; + psample_sample_packet(psample_group, skb, size, iif, oif, + s->rate); + + if (skb_at_tc_ingress(skb) && tcf_sample_dev_ok_push(skb->dev)) + skb_pull(skb, skb->mac_len); + } + + return retval; +} + +static int tcf_sample_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_sample *s = to_sample(a); + struct tc_sample opt = { + .index = s->tcf_index, + .refcnt = refcount_read(&s->tcf_refcnt) - ref, + .bindcnt = atomic_read(&s->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&s->tcf_lock); + opt.action = s->tcf_action; + if (nla_put(skb, TCA_SAMPLE_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + tcf_tm_dump(&t, &s->tcf_tm); + if (nla_put_64bit(skb, TCA_SAMPLE_TM, sizeof(t), &t, TCA_SAMPLE_PAD)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_SAMPLE_RATE, s->rate)) + goto nla_put_failure; + + if (s->truncate) + if (nla_put_u32(skb, TCA_SAMPLE_TRUNC_SIZE, s->trunc_size)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_SAMPLE_PSAMPLE_GROUP, s->psample_group_num)) + goto nla_put_failure; + spin_unlock_bh(&s->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&s->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_sample_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, sample_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_sample_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, sample_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_sample_ops = { + .kind = "sample", + .type = TCA_ACT_SAMPLE, + .owner = THIS_MODULE, + .act = tcf_sample_act, + .dump = tcf_sample_dump, + .init = tcf_sample_init, + .cleanup = tcf_sample_cleanup, + .walk = tcf_sample_walker, + .lookup = tcf_sample_search, + .size = sizeof(struct tcf_sample), +}; + +static __net_init int sample_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, sample_net_id); + + return tc_action_net_init(net, tn, &act_sample_ops); +} + +static void __net_exit sample_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, sample_net_id); +} + +static struct pernet_operations sample_net_ops = { + .init = sample_init_net, + .exit_batch = sample_exit_net, + .id = &sample_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init sample_init_module(void) +{ + return tcf_register_action(&act_sample_ops, &sample_net_ops); +} + +static void __exit sample_cleanup_module(void) +{ + tcf_unregister_action(&act_sample_ops, &sample_net_ops); +} + +module_init(sample_init_module); +module_exit(sample_cleanup_module); + +MODULE_AUTHOR("Yotam Gigi <yotam.gi@gmail.com>"); +MODULE_DESCRIPTION("Packet sampling action"); +MODULE_LICENSE("GPL v2"); diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c new file mode 100644 index 000000000..b418ef62e --- /dev/null +++ b/net/sched/act_simple.c @@ -0,0 +1,251 @@ +/* + * net/sched/act_simple.c Simple example of an action + * + * 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. + * + * Authors: Jamal Hadi Salim (2005-8) + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> + +#define TCA_ACT_SIMP 22 + +#include <linux/tc_act/tc_defact.h> +#include <net/tc_act/tc_defact.h> + +static unsigned int simp_net_id; +static struct tc_action_ops act_simp_ops; + +#define SIMP_MAX_DATA 32 +static int tcf_simp_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_defact *d = to_defact(a); + + spin_lock(&d->tcf_lock); + tcf_lastuse_update(&d->tcf_tm); + bstats_update(&d->tcf_bstats, skb); + + /* print policy string followed by _ then packet count + * Example if this was the 3rd packet and the string was "hello" + * then it would look like "hello_3" (without quotes) + */ + pr_info("simple: %s_%d\n", + (char *)d->tcfd_defdata, d->tcf_bstats.packets); + spin_unlock(&d->tcf_lock); + return d->tcf_action; +} + +static void tcf_simp_release(struct tc_action *a) +{ + struct tcf_defact *d = to_defact(a); + kfree(d->tcfd_defdata); +} + +static int alloc_defdata(struct tcf_defact *d, const struct nlattr *defdata) +{ + d->tcfd_defdata = kzalloc(SIMP_MAX_DATA, GFP_KERNEL); + if (unlikely(!d->tcfd_defdata)) + return -ENOMEM; + nla_strlcpy(d->tcfd_defdata, defdata, SIMP_MAX_DATA); + return 0; +} + +static void reset_policy(struct tcf_defact *d, const struct nlattr *defdata, + struct tc_defact *p) +{ + spin_lock_bh(&d->tcf_lock); + d->tcf_action = p->action; + memset(d->tcfd_defdata, 0, SIMP_MAX_DATA); + nla_strlcpy(d->tcfd_defdata, defdata, SIMP_MAX_DATA); + spin_unlock_bh(&d->tcf_lock); +} + +static const struct nla_policy simple_policy[TCA_DEF_MAX + 1] = { + [TCA_DEF_PARMS] = { .len = sizeof(struct tc_defact) }, + [TCA_DEF_DATA] = { .type = NLA_STRING, .len = SIMP_MAX_DATA }, +}; + +static int tcf_simp_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, simp_net_id); + struct nlattr *tb[TCA_DEF_MAX + 1]; + struct tc_defact *parm; + struct tcf_defact *d; + bool exists = false; + int ret = 0, err; + u32 index; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_DEF_MAX, nla, simple_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_DEF_PARMS] == NULL) + return -EINVAL; + + parm = nla_data(tb[TCA_DEF_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + if (tb[TCA_DEF_DATA] == NULL) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, + &act_simp_ops, bind, false); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + + d = to_defact(*a); + ret = alloc_defdata(d, tb[TCA_DEF_DATA]); + if (ret < 0) { + tcf_idr_release(*a, bind); + return ret; + } + d->tcf_action = parm->action; + ret = ACT_P_CREATED; + } else { + d = to_defact(*a); + + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + + reset_policy(d, tb[TCA_DEF_DATA], parm); + } + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; +} + +static int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_defact *d = to_defact(a); + struct tc_defact opt = { + .index = d->tcf_index, + .refcnt = refcount_read(&d->tcf_refcnt) - ref, + .bindcnt = atomic_read(&d->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&d->tcf_lock); + opt.action = d->tcf_action; + if (nla_put(skb, TCA_DEF_PARMS, sizeof(opt), &opt) || + nla_put_string(skb, TCA_DEF_DATA, d->tcfd_defdata)) + goto nla_put_failure; + + tcf_tm_dump(&t, &d->tcf_tm); + if (nla_put_64bit(skb, TCA_DEF_TM, sizeof(t), &t, TCA_DEF_PAD)) + goto nla_put_failure; + spin_unlock_bh(&d->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&d->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_simp_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, simp_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_simp_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, simp_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_simp_ops = { + .kind = "simple", + .type = TCA_ACT_SIMP, + .owner = THIS_MODULE, + .act = tcf_simp_act, + .dump = tcf_simp_dump, + .cleanup = tcf_simp_release, + .init = tcf_simp_init, + .walk = tcf_simp_walker, + .lookup = tcf_simp_search, + .size = sizeof(struct tcf_defact), +}; + +static __net_init int simp_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, simp_net_id); + + return tc_action_net_init(net, tn, &act_simp_ops); +} + +static void __net_exit simp_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, simp_net_id); +} + +static struct pernet_operations simp_net_ops = { + .init = simp_init_net, + .exit_batch = simp_exit_net, + .id = &simp_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2005)"); +MODULE_DESCRIPTION("Simple example action"); +MODULE_LICENSE("GPL"); + +static int __init simp_init_module(void) +{ + int ret = tcf_register_action(&act_simp_ops, &simp_net_ops); + if (!ret) + pr_info("Simple TC action Loaded\n"); + return ret; +} + +static void __exit simp_cleanup_module(void) +{ + tcf_unregister_action(&act_simp_ops, &simp_net_ops); +} + +module_init(simp_init_module); +module_exit(simp_cleanup_module); diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c new file mode 100644 index 000000000..06648b799 --- /dev/null +++ b/net/sched/act_skbedit.c @@ -0,0 +1,349 @@ +/* + * Copyright (c) 2008, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, see <http://www.gnu.org/licenses/>. + * + * Author: Alexander Duyck <alexander.h.duyck@intel.com> + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/dsfield.h> + +#include <linux/tc_act/tc_skbedit.h> +#include <net/tc_act/tc_skbedit.h> + +static unsigned int skbedit_net_id; +static struct tc_action_ops act_skbedit_ops; + +static int tcf_skbedit_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_skbedit *d = to_skbedit(a); + struct tcf_skbedit_params *params; + int action; + + tcf_lastuse_update(&d->tcf_tm); + bstats_cpu_update(this_cpu_ptr(d->common.cpu_bstats), skb); + + params = rcu_dereference_bh(d->params); + action = READ_ONCE(d->tcf_action); + + if (params->flags & SKBEDIT_F_PRIORITY) + skb->priority = params->priority; + if (params->flags & SKBEDIT_F_INHERITDSFIELD) { + int wlen = skb_network_offset(skb); + + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + wlen += sizeof(struct iphdr); + if (!pskb_may_pull(skb, wlen)) + goto err; + skb->priority = ipv4_get_dsfield(ip_hdr(skb)) >> 2; + break; + + case htons(ETH_P_IPV6): + wlen += sizeof(struct ipv6hdr); + if (!pskb_may_pull(skb, wlen)) + goto err; + skb->priority = ipv6_get_dsfield(ipv6_hdr(skb)) >> 2; + break; + } + } + if (params->flags & SKBEDIT_F_QUEUE_MAPPING && + skb->dev->real_num_tx_queues > params->queue_mapping) + skb_set_queue_mapping(skb, params->queue_mapping); + if (params->flags & SKBEDIT_F_MARK) { + skb->mark &= ~params->mask; + skb->mark |= params->mark & params->mask; + } + if (params->flags & SKBEDIT_F_PTYPE) + skb->pkt_type = params->ptype; + return action; + +err: + qstats_drop_inc(this_cpu_ptr(d->common.cpu_qstats)); + return TC_ACT_SHOT; +} + +static const struct nla_policy skbedit_policy[TCA_SKBEDIT_MAX + 1] = { + [TCA_SKBEDIT_PARMS] = { .len = sizeof(struct tc_skbedit) }, + [TCA_SKBEDIT_PRIORITY] = { .len = sizeof(u32) }, + [TCA_SKBEDIT_QUEUE_MAPPING] = { .len = sizeof(u16) }, + [TCA_SKBEDIT_MARK] = { .len = sizeof(u32) }, + [TCA_SKBEDIT_PTYPE] = { .len = sizeof(u16) }, + [TCA_SKBEDIT_MASK] = { .len = sizeof(u32) }, + [TCA_SKBEDIT_FLAGS] = { .len = sizeof(u64) }, +}; + +static int tcf_skbedit_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, skbedit_net_id); + struct tcf_skbedit_params *params_old, *params_new; + struct nlattr *tb[TCA_SKBEDIT_MAX + 1]; + struct tc_skbedit *parm; + struct tcf_skbedit *d; + u32 flags = 0, *priority = NULL, *mark = NULL, *mask = NULL; + u16 *queue_mapping = NULL, *ptype = NULL; + bool exists = false; + int ret = 0, err; + u32 index; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_SKBEDIT_MAX, nla, skbedit_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_SKBEDIT_PARMS] == NULL) + return -EINVAL; + + if (tb[TCA_SKBEDIT_PRIORITY] != NULL) { + flags |= SKBEDIT_F_PRIORITY; + priority = nla_data(tb[TCA_SKBEDIT_PRIORITY]); + } + + if (tb[TCA_SKBEDIT_QUEUE_MAPPING] != NULL) { + flags |= SKBEDIT_F_QUEUE_MAPPING; + queue_mapping = nla_data(tb[TCA_SKBEDIT_QUEUE_MAPPING]); + } + + if (tb[TCA_SKBEDIT_PTYPE] != NULL) { + ptype = nla_data(tb[TCA_SKBEDIT_PTYPE]); + if (!skb_pkt_type_ok(*ptype)) + return -EINVAL; + flags |= SKBEDIT_F_PTYPE; + } + + if (tb[TCA_SKBEDIT_MARK] != NULL) { + flags |= SKBEDIT_F_MARK; + mark = nla_data(tb[TCA_SKBEDIT_MARK]); + } + + if (tb[TCA_SKBEDIT_MASK] != NULL) { + flags |= SKBEDIT_F_MASK; + mask = nla_data(tb[TCA_SKBEDIT_MASK]); + } + + if (tb[TCA_SKBEDIT_FLAGS] != NULL) { + u64 *pure_flags = nla_data(tb[TCA_SKBEDIT_FLAGS]); + + if (*pure_flags & SKBEDIT_F_INHERITDSFIELD) + flags |= SKBEDIT_F_INHERITDSFIELD; + } + + parm = nla_data(tb[TCA_SKBEDIT_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + if (!flags) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, + &act_skbedit_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + + d = to_skbedit(*a); + ret = ACT_P_CREATED; + } else { + d = to_skbedit(*a); + if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + } + + ASSERT_RTNL(); + + params_new = kzalloc(sizeof(*params_new), GFP_KERNEL); + if (unlikely(!params_new)) { + tcf_idr_release(*a, bind); + return -ENOMEM; + } + + params_new->flags = flags; + if (flags & SKBEDIT_F_PRIORITY) + params_new->priority = *priority; + if (flags & SKBEDIT_F_QUEUE_MAPPING) + params_new->queue_mapping = *queue_mapping; + if (flags & SKBEDIT_F_MARK) + params_new->mark = *mark; + if (flags & SKBEDIT_F_PTYPE) + params_new->ptype = *ptype; + /* default behaviour is to use all the bits */ + params_new->mask = 0xffffffff; + if (flags & SKBEDIT_F_MASK) + params_new->mask = *mask; + + d->tcf_action = parm->action; + params_old = rtnl_dereference(d->params); + rcu_assign_pointer(d->params, params_new); + if (params_old) + kfree_rcu(params_old, rcu); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; +} + +static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_skbedit *d = to_skbedit(a); + struct tcf_skbedit_params *params; + struct tc_skbedit opt = { + .index = d->tcf_index, + .refcnt = refcount_read(&d->tcf_refcnt) - ref, + .bindcnt = atomic_read(&d->tcf_bindcnt) - bind, + .action = d->tcf_action, + }; + u64 pure_flags = 0; + struct tcf_t t; + + params = rtnl_dereference(d->params); + + if (nla_put(skb, TCA_SKBEDIT_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + if ((params->flags & SKBEDIT_F_PRIORITY) && + nla_put_u32(skb, TCA_SKBEDIT_PRIORITY, params->priority)) + goto nla_put_failure; + if ((params->flags & SKBEDIT_F_QUEUE_MAPPING) && + nla_put_u16(skb, TCA_SKBEDIT_QUEUE_MAPPING, params->queue_mapping)) + goto nla_put_failure; + if ((params->flags & SKBEDIT_F_MARK) && + nla_put_u32(skb, TCA_SKBEDIT_MARK, params->mark)) + goto nla_put_failure; + if ((params->flags & SKBEDIT_F_PTYPE) && + nla_put_u16(skb, TCA_SKBEDIT_PTYPE, params->ptype)) + goto nla_put_failure; + if ((params->flags & SKBEDIT_F_MASK) && + nla_put_u32(skb, TCA_SKBEDIT_MASK, params->mask)) + goto nla_put_failure; + if (params->flags & SKBEDIT_F_INHERITDSFIELD) + pure_flags |= SKBEDIT_F_INHERITDSFIELD; + if (pure_flags != 0 && + nla_put(skb, TCA_SKBEDIT_FLAGS, sizeof(pure_flags), &pure_flags)) + goto nla_put_failure; + + tcf_tm_dump(&t, &d->tcf_tm); + if (nla_put_64bit(skb, TCA_SKBEDIT_TM, sizeof(t), &t, TCA_SKBEDIT_PAD)) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static void tcf_skbedit_cleanup(struct tc_action *a) +{ + struct tcf_skbedit *d = to_skbedit(a); + struct tcf_skbedit_params *params; + + params = rcu_dereference_protected(d->params, 1); + if (params) + kfree_rcu(params, rcu); +} + +static int tcf_skbedit_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, skbedit_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_skbedit_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, skbedit_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_skbedit_ops = { + .kind = "skbedit", + .type = TCA_ACT_SKBEDIT, + .owner = THIS_MODULE, + .act = tcf_skbedit_act, + .dump = tcf_skbedit_dump, + .init = tcf_skbedit_init, + .cleanup = tcf_skbedit_cleanup, + .walk = tcf_skbedit_walker, + .lookup = tcf_skbedit_search, + .size = sizeof(struct tcf_skbedit), +}; + +static __net_init int skbedit_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, skbedit_net_id); + + return tc_action_net_init(net, tn, &act_skbedit_ops); +} + +static void __net_exit skbedit_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, skbedit_net_id); +} + +static struct pernet_operations skbedit_net_ops = { + .init = skbedit_init_net, + .exit_batch = skbedit_exit_net, + .id = &skbedit_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Alexander Duyck, <alexander.h.duyck@intel.com>"); +MODULE_DESCRIPTION("SKB Editing"); +MODULE_LICENSE("GPL"); + +static int __init skbedit_init_module(void) +{ + return tcf_register_action(&act_skbedit_ops, &skbedit_net_ops); +} + +static void __exit skbedit_cleanup_module(void) +{ + tcf_unregister_action(&act_skbedit_ops, &skbedit_net_ops); +} + +module_init(skbedit_init_module); +module_exit(skbedit_cleanup_module); diff --git a/net/sched/act_skbmod.c b/net/sched/act_skbmod.c new file mode 100644 index 000000000..03a272af6 --- /dev/null +++ b/net/sched/act_skbmod.c @@ -0,0 +1,314 @@ +/* + * net/sched/act_skbmod.c skb data modifier + * + * Copyright (c) 2016 Jamal Hadi Salim <jhs@mojatatu.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. +*/ + +#include <linux/module.h> +#include <linux/if_arp.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> + +#include <linux/tc_act/tc_skbmod.h> +#include <net/tc_act/tc_skbmod.h> + +static unsigned int skbmod_net_id; +static struct tc_action_ops act_skbmod_ops; + +#define MAX_EDIT_LEN ETH_HLEN +static int tcf_skbmod_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_skbmod *d = to_skbmod(a); + int action; + struct tcf_skbmod_params *p; + u64 flags; + int err; + + tcf_lastuse_update(&d->tcf_tm); + bstats_cpu_update(this_cpu_ptr(d->common.cpu_bstats), skb); + + action = READ_ONCE(d->tcf_action); + if (unlikely(action == TC_ACT_SHOT)) + goto drop; + + if (!skb->dev || skb->dev->type != ARPHRD_ETHER) + return action; + + /* XXX: if you are going to edit more fields beyond ethernet header + * (example when you add IP header replacement or vlan swap) + * then MAX_EDIT_LEN needs to change appropriately + */ + err = skb_ensure_writable(skb, MAX_EDIT_LEN); + if (unlikely(err)) /* best policy is to drop on the floor */ + goto drop; + + p = rcu_dereference_bh(d->skbmod_p); + flags = p->flags; + if (flags & SKBMOD_F_DMAC) + ether_addr_copy(eth_hdr(skb)->h_dest, p->eth_dst); + if (flags & SKBMOD_F_SMAC) + ether_addr_copy(eth_hdr(skb)->h_source, p->eth_src); + if (flags & SKBMOD_F_ETYPE) + eth_hdr(skb)->h_proto = p->eth_type; + + if (flags & SKBMOD_F_SWAPMAC) { + u16 tmpaddr[ETH_ALEN / 2]; /* ether_addr_copy() requirement */ + /*XXX: I am sure we can come up with more efficient swapping*/ + ether_addr_copy((u8 *)tmpaddr, eth_hdr(skb)->h_dest); + ether_addr_copy(eth_hdr(skb)->h_dest, eth_hdr(skb)->h_source); + ether_addr_copy(eth_hdr(skb)->h_source, (u8 *)tmpaddr); + } + + return action; + +drop: + qstats_overlimit_inc(this_cpu_ptr(d->common.cpu_qstats)); + return TC_ACT_SHOT; +} + +static const struct nla_policy skbmod_policy[TCA_SKBMOD_MAX + 1] = { + [TCA_SKBMOD_PARMS] = { .len = sizeof(struct tc_skbmod) }, + [TCA_SKBMOD_DMAC] = { .len = ETH_ALEN }, + [TCA_SKBMOD_SMAC] = { .len = ETH_ALEN }, + [TCA_SKBMOD_ETYPE] = { .type = NLA_U16 }, +}; + +static int tcf_skbmod_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, skbmod_net_id); + struct nlattr *tb[TCA_SKBMOD_MAX + 1]; + struct tcf_skbmod_params *p, *p_old; + struct tc_skbmod *parm; + u32 lflags = 0, index; + struct tcf_skbmod *d; + bool exists = false; + u8 *daddr = NULL; + u8 *saddr = NULL; + u16 eth_type = 0; + int ret = 0, err; + + if (!nla) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_SKBMOD_MAX, nla, skbmod_policy, NULL); + if (err < 0) + return err; + + if (!tb[TCA_SKBMOD_PARMS]) + return -EINVAL; + + if (tb[TCA_SKBMOD_DMAC]) { + daddr = nla_data(tb[TCA_SKBMOD_DMAC]); + lflags |= SKBMOD_F_DMAC; + } + + if (tb[TCA_SKBMOD_SMAC]) { + saddr = nla_data(tb[TCA_SKBMOD_SMAC]); + lflags |= SKBMOD_F_SMAC; + } + + if (tb[TCA_SKBMOD_ETYPE]) { + eth_type = nla_get_u16(tb[TCA_SKBMOD_ETYPE]); + lflags |= SKBMOD_F_ETYPE; + } + + parm = nla_data(tb[TCA_SKBMOD_PARMS]); + index = parm->index; + if (parm->flags & SKBMOD_F_SWAPMAC) + lflags = SKBMOD_F_SWAPMAC; + + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + if (!lflags) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, + &act_skbmod_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + + ret = ACT_P_CREATED; + } else if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + + d = to_skbmod(*a); + + p = kzalloc(sizeof(struct tcf_skbmod_params), GFP_KERNEL); + if (unlikely(!p)) { + tcf_idr_release(*a, bind); + return -ENOMEM; + } + + p->flags = lflags; + d->tcf_action = parm->action; + + if (ovr) + spin_lock_bh(&d->tcf_lock); + /* Protected by tcf_lock if overwriting existing action. */ + p_old = rcu_dereference_protected(d->skbmod_p, 1); + + if (lflags & SKBMOD_F_DMAC) + ether_addr_copy(p->eth_dst, daddr); + if (lflags & SKBMOD_F_SMAC) + ether_addr_copy(p->eth_src, saddr); + if (lflags & SKBMOD_F_ETYPE) + p->eth_type = htons(eth_type); + + rcu_assign_pointer(d->skbmod_p, p); + if (ovr) + spin_unlock_bh(&d->tcf_lock); + + if (p_old) + kfree_rcu(p_old, rcu); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; +} + +static void tcf_skbmod_cleanup(struct tc_action *a) +{ + struct tcf_skbmod *d = to_skbmod(a); + struct tcf_skbmod_params *p; + + p = rcu_dereference_protected(d->skbmod_p, 1); + if (p) + kfree_rcu(p, rcu); +} + +static int tcf_skbmod_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + struct tcf_skbmod *d = to_skbmod(a); + unsigned char *b = skb_tail_pointer(skb); + struct tcf_skbmod_params *p; + struct tc_skbmod opt = { + .index = d->tcf_index, + .refcnt = refcount_read(&d->tcf_refcnt) - ref, + .bindcnt = atomic_read(&d->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&d->tcf_lock); + opt.action = d->tcf_action; + p = rcu_dereference_protected(d->skbmod_p, + lockdep_is_held(&d->tcf_lock)); + opt.flags = p->flags; + if (nla_put(skb, TCA_SKBMOD_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + if ((p->flags & SKBMOD_F_DMAC) && + nla_put(skb, TCA_SKBMOD_DMAC, ETH_ALEN, p->eth_dst)) + goto nla_put_failure; + if ((p->flags & SKBMOD_F_SMAC) && + nla_put(skb, TCA_SKBMOD_SMAC, ETH_ALEN, p->eth_src)) + goto nla_put_failure; + if ((p->flags & SKBMOD_F_ETYPE) && + nla_put_u16(skb, TCA_SKBMOD_ETYPE, ntohs(p->eth_type))) + goto nla_put_failure; + + tcf_tm_dump(&t, &d->tcf_tm); + if (nla_put_64bit(skb, TCA_SKBMOD_TM, sizeof(t), &t, TCA_SKBMOD_PAD)) + goto nla_put_failure; + + spin_unlock_bh(&d->tcf_lock); + return skb->len; +nla_put_failure: + spin_unlock_bh(&d->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_skbmod_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, skbmod_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_skbmod_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, skbmod_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_skbmod_ops = { + .kind = "skbmod", + .type = TCA_ACT_SKBMOD, + .owner = THIS_MODULE, + .act = tcf_skbmod_act, + .dump = tcf_skbmod_dump, + .init = tcf_skbmod_init, + .cleanup = tcf_skbmod_cleanup, + .walk = tcf_skbmod_walker, + .lookup = tcf_skbmod_search, + .size = sizeof(struct tcf_skbmod), +}; + +static __net_init int skbmod_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, skbmod_net_id); + + return tc_action_net_init(net, tn, &act_skbmod_ops); +} + +static void __net_exit skbmod_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, skbmod_net_id); +} + +static struct pernet_operations skbmod_net_ops = { + .init = skbmod_init_net, + .exit_batch = skbmod_exit_net, + .id = &skbmod_net_id, + .size = sizeof(struct tc_action_net), +}; + +MODULE_AUTHOR("Jamal Hadi Salim, <jhs@mojatatu.com>"); +MODULE_DESCRIPTION("SKB data mod-ing"); +MODULE_LICENSE("GPL"); + +static int __init skbmod_init_module(void) +{ + return tcf_register_action(&act_skbmod_ops, &skbmod_net_ops); +} + +static void __exit skbmod_cleanup_module(void) +{ + tcf_unregister_action(&act_skbmod_ops, &skbmod_net_ops); +} + +module_init(skbmod_init_module); +module_exit(skbmod_cleanup_module); diff --git a/net/sched/act_tunnel_key.c b/net/sched/act_tunnel_key.c new file mode 100644 index 000000000..f43234be5 --- /dev/null +++ b/net/sched/act_tunnel_key.c @@ -0,0 +1,616 @@ +/* + * Copyright (c) 2016, Amir Vadai <amir@vadai.me> + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <net/geneve.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/dst.h> + +#include <linux/tc_act/tc_tunnel_key.h> +#include <net/tc_act/tc_tunnel_key.h> + +static unsigned int tunnel_key_net_id; +static struct tc_action_ops act_tunnel_key_ops; + +static int tunnel_key_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_tunnel_key *t = to_tunnel_key(a); + struct tcf_tunnel_key_params *params; + int action; + + params = rcu_dereference_bh(t->params); + + tcf_lastuse_update(&t->tcf_tm); + bstats_cpu_update(this_cpu_ptr(t->common.cpu_bstats), skb); + action = READ_ONCE(t->tcf_action); + + switch (params->tcft_action) { + case TCA_TUNNEL_KEY_ACT_RELEASE: + skb_dst_drop(skb); + break; + case TCA_TUNNEL_KEY_ACT_SET: + skb_dst_drop(skb); + skb_dst_set(skb, dst_clone(¶ms->tcft_enc_metadata->dst)); + break; + default: + WARN_ONCE(1, "Bad tunnel_key action %d.\n", + params->tcft_action); + break; + } + + return action; +} + +static const struct nla_policy +enc_opts_policy[TCA_TUNNEL_KEY_ENC_OPTS_MAX + 1] = { + [TCA_TUNNEL_KEY_ENC_OPTS_GENEVE] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy +geneve_opt_policy[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_MAX + 1] = { + [TCA_TUNNEL_KEY_ENC_OPT_GENEVE_CLASS] = { .type = NLA_U16 }, + [TCA_TUNNEL_KEY_ENC_OPT_GENEVE_TYPE] = { .type = NLA_U8 }, + [TCA_TUNNEL_KEY_ENC_OPT_GENEVE_DATA] = { .type = NLA_BINARY, + .len = 128 }, +}; + +static int +tunnel_key_copy_geneve_opt(const struct nlattr *nla, void *dst, int dst_len, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_MAX + 1]; + int err, data_len, opt_len; + u8 *data; + + err = nla_parse_nested(tb, TCA_TUNNEL_KEY_ENC_OPT_GENEVE_MAX, + nla, geneve_opt_policy, extack); + if (err < 0) + return err; + + if (!tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_CLASS] || + !tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_TYPE] || + !tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_DATA]) { + NL_SET_ERR_MSG(extack, "Missing tunnel key geneve option class, type or data"); + return -EINVAL; + } + + data = nla_data(tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_DATA]); + data_len = nla_len(tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_DATA]); + if (data_len < 4) { + NL_SET_ERR_MSG(extack, "Tunnel key geneve option data is less than 4 bytes long"); + return -ERANGE; + } + if (data_len % 4) { + NL_SET_ERR_MSG(extack, "Tunnel key geneve option data is not a multiple of 4 bytes long"); + return -ERANGE; + } + + opt_len = sizeof(struct geneve_opt) + data_len; + if (dst) { + struct geneve_opt *opt = dst; + + WARN_ON(dst_len < opt_len); + + opt->opt_class = + nla_get_be16(tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_CLASS]); + opt->type = nla_get_u8(tb[TCA_TUNNEL_KEY_ENC_OPT_GENEVE_TYPE]); + opt->length = data_len / 4; /* length is in units of 4 bytes */ + opt->r1 = 0; + opt->r2 = 0; + opt->r3 = 0; + + memcpy(opt + 1, data, data_len); + } + + return opt_len; +} + +static int tunnel_key_copy_opts(const struct nlattr *nla, u8 *dst, + int dst_len, struct netlink_ext_ack *extack) +{ + int err, rem, opt_len, len = nla_len(nla), opts_len = 0; + const struct nlattr *attr, *head = nla_data(nla); + + err = nla_validate(head, len, TCA_TUNNEL_KEY_ENC_OPTS_MAX, + enc_opts_policy, extack); + if (err) + return err; + + nla_for_each_attr(attr, head, len, rem) { + switch (nla_type(attr)) { + case TCA_TUNNEL_KEY_ENC_OPTS_GENEVE: + opt_len = tunnel_key_copy_geneve_opt(attr, dst, + dst_len, extack); + if (opt_len < 0) + return opt_len; + opts_len += opt_len; + if (opts_len > IP_TUNNEL_OPTS_MAX) { + NL_SET_ERR_MSG(extack, "Tunnel options exceeds max size"); + return -EINVAL; + } + if (dst) { + dst_len -= opt_len; + dst += opt_len; + } + break; + } + } + + if (!opts_len) { + NL_SET_ERR_MSG(extack, "Empty list of tunnel options"); + return -EINVAL; + } + + if (rem > 0) { + NL_SET_ERR_MSG(extack, "Trailing data after parsing tunnel key options attributes"); + return -EINVAL; + } + + return opts_len; +} + +static int tunnel_key_get_opts_len(struct nlattr *nla, + struct netlink_ext_ack *extack) +{ + return tunnel_key_copy_opts(nla, NULL, 0, extack); +} + +static int tunnel_key_opts_set(struct nlattr *nla, struct ip_tunnel_info *info, + int opts_len, struct netlink_ext_ack *extack) +{ + info->options_len = opts_len; + switch (nla_type(nla_data(nla))) { + case TCA_TUNNEL_KEY_ENC_OPTS_GENEVE: +#if IS_ENABLED(CONFIG_INET) + info->key.tun_flags |= TUNNEL_GENEVE_OPT; + return tunnel_key_copy_opts(nla, ip_tunnel_info_opts(info), + opts_len, extack); +#else + return -EAFNOSUPPORT; +#endif + default: + NL_SET_ERR_MSG(extack, "Cannot set tunnel options for unknown tunnel type"); + return -EINVAL; + } +} + +static const struct nla_policy tunnel_key_policy[TCA_TUNNEL_KEY_MAX + 1] = { + [TCA_TUNNEL_KEY_PARMS] = { .len = sizeof(struct tc_tunnel_key) }, + [TCA_TUNNEL_KEY_ENC_IPV4_SRC] = { .type = NLA_U32 }, + [TCA_TUNNEL_KEY_ENC_IPV4_DST] = { .type = NLA_U32 }, + [TCA_TUNNEL_KEY_ENC_IPV6_SRC] = { .len = sizeof(struct in6_addr) }, + [TCA_TUNNEL_KEY_ENC_IPV6_DST] = { .len = sizeof(struct in6_addr) }, + [TCA_TUNNEL_KEY_ENC_KEY_ID] = { .type = NLA_U32 }, + [TCA_TUNNEL_KEY_ENC_DST_PORT] = {.type = NLA_U16}, + [TCA_TUNNEL_KEY_NO_CSUM] = { .type = NLA_U8 }, + [TCA_TUNNEL_KEY_ENC_OPTS] = { .type = NLA_NESTED }, + [TCA_TUNNEL_KEY_ENC_TOS] = { .type = NLA_U8 }, + [TCA_TUNNEL_KEY_ENC_TTL] = { .type = NLA_U8 }, +}; + +static void tunnel_key_release_params(struct tcf_tunnel_key_params *p) +{ + if (!p) + return; + if (p->tcft_action == TCA_TUNNEL_KEY_ACT_SET) + dst_release(&p->tcft_enc_metadata->dst); + kfree_rcu(p, rcu); +} + +static int tunnel_key_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, tunnel_key_net_id); + struct nlattr *tb[TCA_TUNNEL_KEY_MAX + 1]; + struct tcf_tunnel_key_params *params_new; + struct metadata_dst *metadata = NULL; + struct tc_tunnel_key *parm; + struct tcf_tunnel_key *t; + bool exists = false; + __be16 dst_port = 0; + int opts_len = 0; + __be64 key_id; + __be16 flags; + u8 tos, ttl; + int ret = 0; + u32 index; + int err; + + if (!nla) { + NL_SET_ERR_MSG(extack, "Tunnel requires attributes to be passed"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_TUNNEL_KEY_MAX, nla, tunnel_key_policy, + extack); + if (err < 0) { + NL_SET_ERR_MSG(extack, "Failed to parse nested tunnel key attributes"); + return err; + } + + if (!tb[TCA_TUNNEL_KEY_PARMS]) { + NL_SET_ERR_MSG(extack, "Missing tunnel key parameters"); + return -EINVAL; + } + + parm = nla_data(tb[TCA_TUNNEL_KEY_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + switch (parm->t_action) { + case TCA_TUNNEL_KEY_ACT_RELEASE: + break; + case TCA_TUNNEL_KEY_ACT_SET: + if (!tb[TCA_TUNNEL_KEY_ENC_KEY_ID]) { + NL_SET_ERR_MSG(extack, "Missing tunnel key id"); + ret = -EINVAL; + goto err_out; + } + + key_id = key32_to_tunnel_id(nla_get_be32(tb[TCA_TUNNEL_KEY_ENC_KEY_ID])); + + flags = TUNNEL_KEY | TUNNEL_CSUM; + if (tb[TCA_TUNNEL_KEY_NO_CSUM] && + nla_get_u8(tb[TCA_TUNNEL_KEY_NO_CSUM])) + flags &= ~TUNNEL_CSUM; + + if (tb[TCA_TUNNEL_KEY_ENC_DST_PORT]) + dst_port = nla_get_be16(tb[TCA_TUNNEL_KEY_ENC_DST_PORT]); + + if (tb[TCA_TUNNEL_KEY_ENC_OPTS]) { + opts_len = tunnel_key_get_opts_len(tb[TCA_TUNNEL_KEY_ENC_OPTS], + extack); + if (opts_len < 0) { + ret = opts_len; + goto err_out; + } + } + + tos = 0; + if (tb[TCA_TUNNEL_KEY_ENC_TOS]) + tos = nla_get_u8(tb[TCA_TUNNEL_KEY_ENC_TOS]); + ttl = 0; + if (tb[TCA_TUNNEL_KEY_ENC_TTL]) + ttl = nla_get_u8(tb[TCA_TUNNEL_KEY_ENC_TTL]); + + if (tb[TCA_TUNNEL_KEY_ENC_IPV4_SRC] && + tb[TCA_TUNNEL_KEY_ENC_IPV4_DST]) { + __be32 saddr; + __be32 daddr; + + saddr = nla_get_in_addr(tb[TCA_TUNNEL_KEY_ENC_IPV4_SRC]); + daddr = nla_get_in_addr(tb[TCA_TUNNEL_KEY_ENC_IPV4_DST]); + + metadata = __ip_tun_set_dst(saddr, daddr, tos, ttl, + dst_port, flags, + key_id, opts_len); + } else if (tb[TCA_TUNNEL_KEY_ENC_IPV6_SRC] && + tb[TCA_TUNNEL_KEY_ENC_IPV6_DST]) { + struct in6_addr saddr; + struct in6_addr daddr; + + saddr = nla_get_in6_addr(tb[TCA_TUNNEL_KEY_ENC_IPV6_SRC]); + daddr = nla_get_in6_addr(tb[TCA_TUNNEL_KEY_ENC_IPV6_DST]); + + metadata = __ipv6_tun_set_dst(&saddr, &daddr, tos, ttl, dst_port, + 0, flags, + key_id, opts_len); + } else { + NL_SET_ERR_MSG(extack, "Missing either ipv4 or ipv6 src and dst"); + ret = -EINVAL; + goto err_out; + } + + if (!metadata) { + NL_SET_ERR_MSG(extack, "Cannot allocate tunnel metadata dst"); + ret = -ENOMEM; + goto err_out; + } + + if (opts_len) { + ret = tunnel_key_opts_set(tb[TCA_TUNNEL_KEY_ENC_OPTS], + &metadata->u.tun_info, + opts_len, extack); + if (ret < 0) + goto release_tun_meta; + } + + metadata->u.tun_info.mode |= IP_TUNNEL_INFO_TX; + break; + default: + NL_SET_ERR_MSG(extack, "Unknown tunnel key action"); + ret = -EINVAL; + goto err_out; + } + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, + &act_tunnel_key_ops, bind, true); + if (ret) { + NL_SET_ERR_MSG(extack, "Cannot create TC IDR"); + goto release_tun_meta; + } + + ret = ACT_P_CREATED; + } else if (!ovr) { + NL_SET_ERR_MSG(extack, "TC IDR already exists"); + ret = -EEXIST; + goto release_tun_meta; + } + + t = to_tunnel_key(*a); + + params_new = kzalloc(sizeof(*params_new), GFP_KERNEL); + if (unlikely(!params_new)) { + NL_SET_ERR_MSG(extack, "Cannot allocate tunnel key parameters"); + ret = -ENOMEM; + exists = true; + goto release_tun_meta; + } + params_new->tcft_action = parm->t_action; + params_new->tcft_enc_metadata = metadata; + + spin_lock_bh(&t->tcf_lock); + t->tcf_action = parm->action; + rcu_swap_protected(t->params, params_new, + lockdep_is_held(&t->tcf_lock)); + spin_unlock_bh(&t->tcf_lock); + tunnel_key_release_params(params_new); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + + return ret; + +release_tun_meta: + if (metadata) + dst_release(&metadata->dst); + +err_out: + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return ret; +} + +static void tunnel_key_release(struct tc_action *a) +{ + struct tcf_tunnel_key *t = to_tunnel_key(a); + struct tcf_tunnel_key_params *params; + + params = rcu_dereference_protected(t->params, 1); + tunnel_key_release_params(params); +} + +static int tunnel_key_geneve_opts_dump(struct sk_buff *skb, + const struct ip_tunnel_info *info) +{ + int len = info->options_len; + u8 *src = (u8 *)(info + 1); + struct nlattr *start; + + start = nla_nest_start(skb, TCA_TUNNEL_KEY_ENC_OPTS_GENEVE); + if (!start) + return -EMSGSIZE; + + while (len > 0) { + struct geneve_opt *opt = (struct geneve_opt *)src; + + if (nla_put_be16(skb, TCA_TUNNEL_KEY_ENC_OPT_GENEVE_CLASS, + opt->opt_class) || + nla_put_u8(skb, TCA_TUNNEL_KEY_ENC_OPT_GENEVE_TYPE, + opt->type) || + nla_put(skb, TCA_TUNNEL_KEY_ENC_OPT_GENEVE_DATA, + opt->length * 4, opt + 1)) { + nla_nest_cancel(skb, start); + return -EMSGSIZE; + } + + len -= sizeof(struct geneve_opt) + opt->length * 4; + src += sizeof(struct geneve_opt) + opt->length * 4; + } + + nla_nest_end(skb, start); + return 0; +} + +static int tunnel_key_opts_dump(struct sk_buff *skb, + const struct ip_tunnel_info *info) +{ + struct nlattr *start; + int err = -EINVAL; + + if (!info->options_len) + return 0; + + start = nla_nest_start(skb, TCA_TUNNEL_KEY_ENC_OPTS); + if (!start) + return -EMSGSIZE; + + if (info->key.tun_flags & TUNNEL_GENEVE_OPT) { + err = tunnel_key_geneve_opts_dump(skb, info); + if (err) + goto err_out; + } else { +err_out: + nla_nest_cancel(skb, start); + return err; + } + + nla_nest_end(skb, start); + return 0; +} + +static int tunnel_key_dump_addresses(struct sk_buff *skb, + const struct ip_tunnel_info *info) +{ + unsigned short family = ip_tunnel_info_af(info); + + if (family == AF_INET) { + __be32 saddr = info->key.u.ipv4.src; + __be32 daddr = info->key.u.ipv4.dst; + + if (!nla_put_in_addr(skb, TCA_TUNNEL_KEY_ENC_IPV4_SRC, saddr) && + !nla_put_in_addr(skb, TCA_TUNNEL_KEY_ENC_IPV4_DST, daddr)) + return 0; + } + + if (family == AF_INET6) { + const struct in6_addr *saddr6 = &info->key.u.ipv6.src; + const struct in6_addr *daddr6 = &info->key.u.ipv6.dst; + + if (!nla_put_in6_addr(skb, + TCA_TUNNEL_KEY_ENC_IPV6_SRC, saddr6) && + !nla_put_in6_addr(skb, + TCA_TUNNEL_KEY_ENC_IPV6_DST, daddr6)) + return 0; + } + + return -EINVAL; +} + +static int tunnel_key_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_tunnel_key *t = to_tunnel_key(a); + struct tcf_tunnel_key_params *params; + struct tc_tunnel_key opt = { + .index = t->tcf_index, + .refcnt = refcount_read(&t->tcf_refcnt) - ref, + .bindcnt = atomic_read(&t->tcf_bindcnt) - bind, + }; + struct tcf_t tm; + + spin_lock_bh(&t->tcf_lock); + params = rcu_dereference_protected(t->params, + lockdep_is_held(&t->tcf_lock)); + opt.action = t->tcf_action; + opt.t_action = params->tcft_action; + + if (nla_put(skb, TCA_TUNNEL_KEY_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + if (params->tcft_action == TCA_TUNNEL_KEY_ACT_SET) { + struct ip_tunnel_info *info = + ¶ms->tcft_enc_metadata->u.tun_info; + struct ip_tunnel_key *key = &info->key; + __be32 key_id = tunnel_id_to_key32(key->tun_id); + + if (nla_put_be32(skb, TCA_TUNNEL_KEY_ENC_KEY_ID, key_id) || + tunnel_key_dump_addresses(skb, + ¶ms->tcft_enc_metadata->u.tun_info) || + nla_put_be16(skb, TCA_TUNNEL_KEY_ENC_DST_PORT, key->tp_dst) || + nla_put_u8(skb, TCA_TUNNEL_KEY_NO_CSUM, + !(key->tun_flags & TUNNEL_CSUM)) || + tunnel_key_opts_dump(skb, info)) + goto nla_put_failure; + + if (key->tos && nla_put_u8(skb, TCA_TUNNEL_KEY_ENC_TOS, key->tos)) + goto nla_put_failure; + + if (key->ttl && nla_put_u8(skb, TCA_TUNNEL_KEY_ENC_TTL, key->ttl)) + goto nla_put_failure; + } + + tcf_tm_dump(&tm, &t->tcf_tm); + if (nla_put_64bit(skb, TCA_TUNNEL_KEY_TM, sizeof(tm), + &tm, TCA_TUNNEL_KEY_PAD)) + goto nla_put_failure; + spin_unlock_bh(&t->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&t->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tunnel_key_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, tunnel_key_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tunnel_key_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, tunnel_key_net_id); + + return tcf_idr_search(tn, a, index); +} + +static struct tc_action_ops act_tunnel_key_ops = { + .kind = "tunnel_key", + .type = TCA_ACT_TUNNEL_KEY, + .owner = THIS_MODULE, + .act = tunnel_key_act, + .dump = tunnel_key_dump, + .init = tunnel_key_init, + .cleanup = tunnel_key_release, + .walk = tunnel_key_walker, + .lookup = tunnel_key_search, + .size = sizeof(struct tcf_tunnel_key), +}; + +static __net_init int tunnel_key_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, tunnel_key_net_id); + + return tc_action_net_init(net, tn, &act_tunnel_key_ops); +} + +static void __net_exit tunnel_key_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, tunnel_key_net_id); +} + +static struct pernet_operations tunnel_key_net_ops = { + .init = tunnel_key_init_net, + .exit_batch = tunnel_key_exit_net, + .id = &tunnel_key_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init tunnel_key_init_module(void) +{ + return tcf_register_action(&act_tunnel_key_ops, &tunnel_key_net_ops); +} + +static void __exit tunnel_key_cleanup_module(void) +{ + tcf_unregister_action(&act_tunnel_key_ops, &tunnel_key_net_ops); +} + +module_init(tunnel_key_init_module); +module_exit(tunnel_key_cleanup_module); + +MODULE_AUTHOR("Amir Vadai <amir@vadai.me>"); +MODULE_DESCRIPTION("ip tunnel manipulation actions"); +MODULE_LICENSE("GPL v2"); diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c new file mode 100644 index 000000000..41528b966 --- /dev/null +++ b/net/sched/act_vlan.c @@ -0,0 +1,357 @@ +/* + * Copyright (c) 2014 Jiri Pirko <jiri@resnulli.us> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/if_vlan.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> + +#include <linux/tc_act/tc_vlan.h> +#include <net/tc_act/tc_vlan.h> + +static unsigned int vlan_net_id; +static struct tc_action_ops act_vlan_ops; + +static int tcf_vlan_act(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + struct tcf_vlan *v = to_vlan(a); + struct tcf_vlan_params *p; + int action; + int err; + u16 tci; + + tcf_lastuse_update(&v->tcf_tm); + bstats_cpu_update(this_cpu_ptr(v->common.cpu_bstats), skb); + + /* Ensure 'data' points at mac_header prior calling vlan manipulating + * functions. + */ + if (skb_at_tc_ingress(skb)) + skb_push_rcsum(skb, skb->mac_len); + + action = READ_ONCE(v->tcf_action); + + p = rcu_dereference_bh(v->vlan_p); + + switch (p->tcfv_action) { + case TCA_VLAN_ACT_POP: + err = skb_vlan_pop(skb); + if (err) + goto drop; + break; + case TCA_VLAN_ACT_PUSH: + err = skb_vlan_push(skb, p->tcfv_push_proto, p->tcfv_push_vid | + (p->tcfv_push_prio << VLAN_PRIO_SHIFT)); + if (err) + goto drop; + break; + case TCA_VLAN_ACT_MODIFY: + /* No-op if no vlan tag (either hw-accel or in-payload) */ + if (!skb_vlan_tagged(skb)) + goto out; + /* extract existing tag (and guarantee no hw-accel tag) */ + if (skb_vlan_tag_present(skb)) { + tci = skb_vlan_tag_get(skb); + skb->vlan_tci = 0; + } else { + /* in-payload vlan tag, pop it */ + err = __skb_vlan_pop(skb, &tci); + if (err) + goto drop; + } + /* replace the vid */ + tci = (tci & ~VLAN_VID_MASK) | p->tcfv_push_vid; + /* replace prio bits, if tcfv_push_prio specified */ + if (p->tcfv_push_prio) { + tci &= ~VLAN_PRIO_MASK; + tci |= p->tcfv_push_prio << VLAN_PRIO_SHIFT; + } + /* put updated tci as hwaccel tag */ + __vlan_hwaccel_put_tag(skb, p->tcfv_push_proto, tci); + break; + default: + BUG(); + } + +out: + if (skb_at_tc_ingress(skb)) + skb_pull_rcsum(skb, skb->mac_len); + + return action; + +drop: + qstats_drop_inc(this_cpu_ptr(v->common.cpu_qstats)); + return TC_ACT_SHOT; +} + +static const struct nla_policy vlan_policy[TCA_VLAN_MAX + 1] = { + [TCA_VLAN_PARMS] = { .len = sizeof(struct tc_vlan) }, + [TCA_VLAN_PUSH_VLAN_ID] = { .type = NLA_U16 }, + [TCA_VLAN_PUSH_VLAN_PROTOCOL] = { .type = NLA_U16 }, + [TCA_VLAN_PUSH_VLAN_PRIORITY] = { .type = NLA_U8 }, +}; + +static int tcf_vlan_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action **a, + int ovr, int bind, bool rtnl_held, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, vlan_net_id); + struct nlattr *tb[TCA_VLAN_MAX + 1]; + struct tcf_vlan_params *p; + struct tc_vlan *parm; + struct tcf_vlan *v; + int action; + u16 push_vid = 0; + __be16 push_proto = 0; + u8 push_prio = 0; + bool exists = false; + int ret = 0, err; + u32 index; + + if (!nla) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_VLAN_MAX, nla, vlan_policy, NULL); + if (err < 0) + return err; + + if (!tb[TCA_VLAN_PARMS]) + return -EINVAL; + parm = nla_data(tb[TCA_VLAN_PARMS]); + index = parm->index; + err = tcf_idr_check_alloc(tn, &index, a, bind); + if (err < 0) + return err; + exists = err; + if (exists && bind) + return 0; + + switch (parm->v_action) { + case TCA_VLAN_ACT_POP: + break; + case TCA_VLAN_ACT_PUSH: + case TCA_VLAN_ACT_MODIFY: + if (!tb[TCA_VLAN_PUSH_VLAN_ID]) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + push_vid = nla_get_u16(tb[TCA_VLAN_PUSH_VLAN_ID]); + if (push_vid >= VLAN_VID_MASK) { + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -ERANGE; + } + + if (tb[TCA_VLAN_PUSH_VLAN_PROTOCOL]) { + push_proto = nla_get_be16(tb[TCA_VLAN_PUSH_VLAN_PROTOCOL]); + switch (push_proto) { + case htons(ETH_P_8021Q): + case htons(ETH_P_8021AD): + break; + default: + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EPROTONOSUPPORT; + } + } else { + push_proto = htons(ETH_P_8021Q); + } + + if (tb[TCA_VLAN_PUSH_VLAN_PRIORITY]) + push_prio = nla_get_u8(tb[TCA_VLAN_PUSH_VLAN_PRIORITY]); + break; + default: + if (exists) + tcf_idr_release(*a, bind); + else + tcf_idr_cleanup(tn, index); + return -EINVAL; + } + action = parm->v_action; + + if (!exists) { + ret = tcf_idr_create(tn, index, est, a, + &act_vlan_ops, bind, true); + if (ret) { + tcf_idr_cleanup(tn, index); + return ret; + } + + ret = ACT_P_CREATED; + } else if (!ovr) { + tcf_idr_release(*a, bind); + return -EEXIST; + } + + v = to_vlan(*a); + + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) { + tcf_idr_release(*a, bind); + return -ENOMEM; + } + + p->tcfv_action = action; + p->tcfv_push_vid = push_vid; + p->tcfv_push_prio = push_prio; + p->tcfv_push_proto = push_proto; + + spin_lock_bh(&v->tcf_lock); + v->tcf_action = parm->action; + rcu_swap_protected(v->vlan_p, p, lockdep_is_held(&v->tcf_lock)); + spin_unlock_bh(&v->tcf_lock); + + if (p) + kfree_rcu(p, rcu); + + if (ret == ACT_P_CREATED) + tcf_idr_insert(tn, *a); + return ret; +} + +static void tcf_vlan_cleanup(struct tc_action *a) +{ + struct tcf_vlan *v = to_vlan(a); + struct tcf_vlan_params *p; + + p = rcu_dereference_protected(v->vlan_p, 1); + if (p) + kfree_rcu(p, rcu); +} + +static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_vlan *v = to_vlan(a); + struct tcf_vlan_params *p; + struct tc_vlan opt = { + .index = v->tcf_index, + .refcnt = refcount_read(&v->tcf_refcnt) - ref, + .bindcnt = atomic_read(&v->tcf_bindcnt) - bind, + }; + struct tcf_t t; + + spin_lock_bh(&v->tcf_lock); + opt.action = v->tcf_action; + p = rcu_dereference_protected(v->vlan_p, lockdep_is_held(&v->tcf_lock)); + opt.v_action = p->tcfv_action; + if (nla_put(skb, TCA_VLAN_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + if ((p->tcfv_action == TCA_VLAN_ACT_PUSH || + p->tcfv_action == TCA_VLAN_ACT_MODIFY) && + (nla_put_u16(skb, TCA_VLAN_PUSH_VLAN_ID, p->tcfv_push_vid) || + nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, + p->tcfv_push_proto) || + (nla_put_u8(skb, TCA_VLAN_PUSH_VLAN_PRIORITY, + p->tcfv_push_prio)))) + goto nla_put_failure; + + tcf_tm_dump(&t, &v->tcf_tm); + if (nla_put_64bit(skb, TCA_VLAN_TM, sizeof(t), &t, TCA_VLAN_PAD)) + goto nla_put_failure; + spin_unlock_bh(&v->tcf_lock); + + return skb->len; + +nla_put_failure: + spin_unlock_bh(&v->tcf_lock); + nlmsg_trim(skb, b); + return -1; +} + +static int tcf_vlan_walker(struct net *net, struct sk_buff *skb, + struct netlink_callback *cb, int type, + const struct tc_action_ops *ops, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, vlan_net_id); + + return tcf_generic_walker(tn, skb, cb, type, ops, extack); +} + +static int tcf_vlan_search(struct net *net, struct tc_action **a, u32 index, + struct netlink_ext_ack *extack) +{ + struct tc_action_net *tn = net_generic(net, vlan_net_id); + + return tcf_idr_search(tn, a, index); +} + +static size_t tcf_vlan_get_fill_size(const struct tc_action *act) +{ + return nla_total_size(sizeof(struct tc_vlan)) + + nla_total_size(sizeof(u16)) /* TCA_VLAN_PUSH_VLAN_ID */ + + nla_total_size(sizeof(u16)) /* TCA_VLAN_PUSH_VLAN_PROTOCOL */ + + nla_total_size(sizeof(u8)); /* TCA_VLAN_PUSH_VLAN_PRIORITY */ +} + +static struct tc_action_ops act_vlan_ops = { + .kind = "vlan", + .type = TCA_ACT_VLAN, + .owner = THIS_MODULE, + .act = tcf_vlan_act, + .dump = tcf_vlan_dump, + .init = tcf_vlan_init, + .cleanup = tcf_vlan_cleanup, + .walk = tcf_vlan_walker, + .get_fill_size = tcf_vlan_get_fill_size, + .lookup = tcf_vlan_search, + .size = sizeof(struct tcf_vlan), +}; + +static __net_init int vlan_init_net(struct net *net) +{ + struct tc_action_net *tn = net_generic(net, vlan_net_id); + + return tc_action_net_init(net, tn, &act_vlan_ops); +} + +static void __net_exit vlan_exit_net(struct list_head *net_list) +{ + tc_action_net_exit(net_list, vlan_net_id); +} + +static struct pernet_operations vlan_net_ops = { + .init = vlan_init_net, + .exit_batch = vlan_exit_net, + .id = &vlan_net_id, + .size = sizeof(struct tc_action_net), +}; + +static int __init vlan_init_module(void) +{ + return tcf_register_action(&act_vlan_ops, &vlan_net_ops); +} + +static void __exit vlan_cleanup_module(void) +{ + tcf_unregister_action(&act_vlan_ops, &vlan_net_ops); +} + +module_init(vlan_init_module); +module_exit(vlan_cleanup_module); + +MODULE_AUTHOR("Jiri Pirko <jiri@resnulli.us>"); +MODULE_DESCRIPTION("vlan manipulation actions"); +MODULE_LICENSE("GPL v2"); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c new file mode 100644 index 000000000..435911dc9 --- /dev/null +++ b/net/sched/cls_api.c @@ -0,0 +1,2320 @@ +/* + * net/sched/cls_api.c Packet classifier API. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + * Changes: + * + * Eduardo J. Blanco <ejbs@netlabs.com.uy> :990222: kmod support + * + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/err.h> +#include <linux/skbuff.h> +#include <linux/init.h> +#include <linux/kmod.h> +#include <linux/slab.h> +#include <linux/idr.h> +#include <net/net_namespace.h> +#include <net/sock.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +extern const struct nla_policy rtm_tca_policy[TCA_MAX + 1]; + +/* The list of all installed classifier types */ +static LIST_HEAD(tcf_proto_base); + +/* Protects list of registered TC modules. It is pure SMP lock. */ +static DEFINE_RWLOCK(cls_mod_lock); + +/* Find classifier type by string name */ + +static const struct tcf_proto_ops *__tcf_proto_lookup_ops(const char *kind) +{ + const struct tcf_proto_ops *t, *res = NULL; + + if (kind) { + read_lock(&cls_mod_lock); + list_for_each_entry(t, &tcf_proto_base, head) { + if (strcmp(kind, t->kind) == 0) { + if (try_module_get(t->owner)) + res = t; + break; + } + } + read_unlock(&cls_mod_lock); + } + return res; +} + +static const struct tcf_proto_ops * +tcf_proto_lookup_ops(const char *kind, struct netlink_ext_ack *extack) +{ + const struct tcf_proto_ops *ops; + + ops = __tcf_proto_lookup_ops(kind); + if (ops) + return ops; +#ifdef CONFIG_MODULES + rtnl_unlock(); + request_module("cls_%s", kind); + rtnl_lock(); + ops = __tcf_proto_lookup_ops(kind); + /* We dropped the RTNL semaphore in order to perform + * the module load. So, even if we succeeded in loading + * the module we have to replay the request. We indicate + * this using -EAGAIN. + */ + if (ops) { + module_put(ops->owner); + return ERR_PTR(-EAGAIN); + } +#endif + NL_SET_ERR_MSG(extack, "TC classifier not found"); + return ERR_PTR(-ENOENT); +} + +/* Register(unregister) new classifier type */ + +int register_tcf_proto_ops(struct tcf_proto_ops *ops) +{ + struct tcf_proto_ops *t; + int rc = -EEXIST; + + write_lock(&cls_mod_lock); + list_for_each_entry(t, &tcf_proto_base, head) + if (!strcmp(ops->kind, t->kind)) + goto out; + + list_add_tail(&ops->head, &tcf_proto_base); + rc = 0; +out: + write_unlock(&cls_mod_lock); + return rc; +} +EXPORT_SYMBOL(register_tcf_proto_ops); + +static struct workqueue_struct *tc_filter_wq; + +int unregister_tcf_proto_ops(struct tcf_proto_ops *ops) +{ + struct tcf_proto_ops *t; + int rc = -ENOENT; + + /* Wait for outstanding call_rcu()s, if any, from a + * tcf_proto_ops's destroy() handler. + */ + rcu_barrier(); + flush_workqueue(tc_filter_wq); + + write_lock(&cls_mod_lock); + list_for_each_entry(t, &tcf_proto_base, head) { + if (t == ops) { + list_del(&t->head); + rc = 0; + break; + } + } + write_unlock(&cls_mod_lock); + return rc; +} +EXPORT_SYMBOL(unregister_tcf_proto_ops); + +bool tcf_queue_work(struct rcu_work *rwork, work_func_t func) +{ + INIT_RCU_WORK(rwork, func); + return queue_rcu_work(tc_filter_wq, rwork); +} +EXPORT_SYMBOL(tcf_queue_work); + +/* Select new prio value from the range, managed by kernel. */ + +static inline u32 tcf_auto_prio(struct tcf_proto *tp) +{ + u32 first = TC_H_MAKE(0xC0000000U, 0U); + + if (tp) + first = tp->prio - 1; + + return TC_H_MAJ(first); +} + +static struct tcf_proto *tcf_proto_create(const char *kind, u32 protocol, + u32 prio, struct tcf_chain *chain, + struct netlink_ext_ack *extack) +{ + struct tcf_proto *tp; + int err; + + tp = kzalloc(sizeof(*tp), GFP_KERNEL); + if (!tp) + return ERR_PTR(-ENOBUFS); + + tp->ops = tcf_proto_lookup_ops(kind, extack); + if (IS_ERR(tp->ops)) { + err = PTR_ERR(tp->ops); + goto errout; + } + tp->classify = tp->ops->classify; + tp->protocol = protocol; + tp->prio = prio; + tp->chain = chain; + + err = tp->ops->init(tp); + if (err) { + module_put(tp->ops->owner); + goto errout; + } + return tp; + +errout: + kfree(tp); + return ERR_PTR(err); +} + +static void tcf_proto_destroy(struct tcf_proto *tp, + struct netlink_ext_ack *extack) +{ + tp->ops->destroy(tp, extack); + module_put(tp->ops->owner); + kfree_rcu(tp, rcu); +} + +struct tcf_filter_chain_list_item { + struct list_head list; + tcf_chain_head_change_t *chain_head_change; + void *chain_head_change_priv; +}; + +static struct tcf_chain *tcf_chain_create(struct tcf_block *block, + u32 chain_index) +{ + struct tcf_chain *chain; + + chain = kzalloc(sizeof(*chain), GFP_KERNEL); + if (!chain) + return NULL; + list_add_tail(&chain->list, &block->chain_list); + chain->block = block; + chain->index = chain_index; + chain->refcnt = 1; + if (!chain->index) + block->chain0.chain = chain; + return chain; +} + +static void tcf_chain_head_change_item(struct tcf_filter_chain_list_item *item, + struct tcf_proto *tp_head) +{ + if (item->chain_head_change) + item->chain_head_change(tp_head, item->chain_head_change_priv); +} + +static void tcf_chain0_head_change(struct tcf_chain *chain, + struct tcf_proto *tp_head) +{ + struct tcf_filter_chain_list_item *item; + struct tcf_block *block = chain->block; + + if (chain->index) + return; + list_for_each_entry(item, &block->chain0.filter_chain_list, list) + tcf_chain_head_change_item(item, tp_head); +} + +static void tcf_chain_destroy(struct tcf_chain *chain) +{ + struct tcf_block *block = chain->block; + + list_del(&chain->list); + if (!chain->index) + block->chain0.chain = NULL; + kfree(chain); + if (list_empty(&block->chain_list) && block->refcnt == 0) + kfree(block); +} + +static void tcf_chain_hold(struct tcf_chain *chain) +{ + ++chain->refcnt; +} + +static bool tcf_chain_held_by_acts_only(struct tcf_chain *chain) +{ + /* In case all the references are action references, this + * chain should not be shown to the user. + */ + return chain->refcnt == chain->action_refcnt; +} + +static struct tcf_chain *tcf_chain_lookup(struct tcf_block *block, + u32 chain_index) +{ + struct tcf_chain *chain; + + list_for_each_entry(chain, &block->chain_list, list) { + if (chain->index == chain_index) + return chain; + } + return NULL; +} + +static int tc_chain_notify(struct tcf_chain *chain, struct sk_buff *oskb, + u32 seq, u16 flags, int event, bool unicast); + +static struct tcf_chain *__tcf_chain_get(struct tcf_block *block, + u32 chain_index, bool create, + bool by_act) +{ + struct tcf_chain *chain = tcf_chain_lookup(block, chain_index); + + if (chain) { + tcf_chain_hold(chain); + } else { + if (!create) + return NULL; + chain = tcf_chain_create(block, chain_index); + if (!chain) + return NULL; + } + + if (by_act) + ++chain->action_refcnt; + + /* Send notification only in case we got the first + * non-action reference. Until then, the chain acts only as + * a placeholder for actions pointing to it and user ought + * not know about them. + */ + if (chain->refcnt - chain->action_refcnt == 1 && !by_act) + tc_chain_notify(chain, NULL, 0, NLM_F_CREATE | NLM_F_EXCL, + RTM_NEWCHAIN, false); + + return chain; +} + +static struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, + bool create) +{ + return __tcf_chain_get(block, chain_index, create, false); +} + +struct tcf_chain *tcf_chain_get_by_act(struct tcf_block *block, u32 chain_index) +{ + return __tcf_chain_get(block, chain_index, true, true); +} +EXPORT_SYMBOL(tcf_chain_get_by_act); + +static void tc_chain_tmplt_del(struct tcf_chain *chain); + +static void __tcf_chain_put(struct tcf_chain *chain, bool by_act) +{ + if (by_act) + chain->action_refcnt--; + chain->refcnt--; + + /* The last dropped non-action reference will trigger notification. */ + if (chain->refcnt - chain->action_refcnt == 0 && !by_act) + tc_chain_notify(chain, NULL, 0, 0, RTM_DELCHAIN, false); + + if (chain->refcnt == 0) { + tc_chain_tmplt_del(chain); + tcf_chain_destroy(chain); + } +} + +static void tcf_chain_put(struct tcf_chain *chain) +{ + __tcf_chain_put(chain, false); +} + +void tcf_chain_put_by_act(struct tcf_chain *chain) +{ + __tcf_chain_put(chain, true); +} +EXPORT_SYMBOL(tcf_chain_put_by_act); + +static void tcf_chain_put_explicitly_created(struct tcf_chain *chain) +{ + if (chain->explicitly_created) + tcf_chain_put(chain); +} + +static void tcf_chain_flush(struct tcf_chain *chain) +{ + struct tcf_proto *tp = rtnl_dereference(chain->filter_chain); + + tcf_chain0_head_change(chain, NULL); + while (tp) { + RCU_INIT_POINTER(chain->filter_chain, tp->next); + tcf_proto_destroy(tp, NULL); + tp = rtnl_dereference(chain->filter_chain); + tcf_chain_put(chain); + } +} + +static bool tcf_block_offload_in_use(struct tcf_block *block) +{ + return block->offloadcnt; +} + +static int tcf_block_offload_cmd(struct tcf_block *block, + struct net_device *dev, + struct tcf_block_ext_info *ei, + enum tc_block_command command, + struct netlink_ext_ack *extack) +{ + struct tc_block_offload bo = {}; + + bo.command = command; + bo.binder_type = ei->binder_type; + bo.block = block; + bo.extack = extack; + return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_BLOCK, &bo); +} + +static int tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q, + struct tcf_block_ext_info *ei, + struct netlink_ext_ack *extack) +{ + struct net_device *dev = q->dev_queue->dev; + int err; + + if (!dev->netdev_ops->ndo_setup_tc) + goto no_offload_dev_inc; + + /* If tc offload feature is disabled and the block we try to bind + * to already has some offloaded filters, forbid to bind. + */ + if (!tc_can_offload(dev) && tcf_block_offload_in_use(block)) { + NL_SET_ERR_MSG(extack, "Bind to offloaded block failed as dev has offload disabled"); + return -EOPNOTSUPP; + } + + err = tcf_block_offload_cmd(block, dev, ei, TC_BLOCK_BIND, extack); + if (err == -EOPNOTSUPP) + goto no_offload_dev_inc; + return err; + +no_offload_dev_inc: + if (tcf_block_offload_in_use(block)) + return -EOPNOTSUPP; + block->nooffloaddevcnt++; + return 0; +} + +static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q, + struct tcf_block_ext_info *ei) +{ + struct net_device *dev = q->dev_queue->dev; + int err; + + if (!dev->netdev_ops->ndo_setup_tc) + goto no_offload_dev_dec; + err = tcf_block_offload_cmd(block, dev, ei, TC_BLOCK_UNBIND, NULL); + if (err == -EOPNOTSUPP) + goto no_offload_dev_dec; + return; + +no_offload_dev_dec: + WARN_ON(block->nooffloaddevcnt-- == 0); +} + +static int +tcf_chain0_head_change_cb_add(struct tcf_block *block, + struct tcf_block_ext_info *ei, + struct netlink_ext_ack *extack) +{ + struct tcf_chain *chain0 = block->chain0.chain; + struct tcf_filter_chain_list_item *item; + + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) { + NL_SET_ERR_MSG(extack, "Memory allocation for head change callback item failed"); + return -ENOMEM; + } + item->chain_head_change = ei->chain_head_change; + item->chain_head_change_priv = ei->chain_head_change_priv; + if (chain0 && chain0->filter_chain) + tcf_chain_head_change_item(item, chain0->filter_chain); + list_add(&item->list, &block->chain0.filter_chain_list); + return 0; +} + +static void +tcf_chain0_head_change_cb_del(struct tcf_block *block, + struct tcf_block_ext_info *ei) +{ + struct tcf_chain *chain0 = block->chain0.chain; + struct tcf_filter_chain_list_item *item; + + list_for_each_entry(item, &block->chain0.filter_chain_list, list) { + if ((!ei->chain_head_change && !ei->chain_head_change_priv) || + (item->chain_head_change == ei->chain_head_change && + item->chain_head_change_priv == ei->chain_head_change_priv)) { + if (chain0) + tcf_chain_head_change_item(item, NULL); + list_del(&item->list); + kfree(item); + return; + } + } + WARN_ON(1); +} + +struct tcf_net { + struct idr idr; +}; + +static unsigned int tcf_net_id; + +static int tcf_block_insert(struct tcf_block *block, struct net *net, + struct netlink_ext_ack *extack) +{ + struct tcf_net *tn = net_generic(net, tcf_net_id); + + return idr_alloc_u32(&tn->idr, block, &block->index, block->index, + GFP_KERNEL); +} + +static void tcf_block_remove(struct tcf_block *block, struct net *net) +{ + struct tcf_net *tn = net_generic(net, tcf_net_id); + + idr_remove(&tn->idr, block->index); +} + +static struct tcf_block *tcf_block_create(struct net *net, struct Qdisc *q, + u32 block_index, + struct netlink_ext_ack *extack) +{ + struct tcf_block *block; + + block = kzalloc(sizeof(*block), GFP_KERNEL); + if (!block) { + NL_SET_ERR_MSG(extack, "Memory allocation for block failed"); + return ERR_PTR(-ENOMEM); + } + INIT_LIST_HEAD(&block->chain_list); + INIT_LIST_HEAD(&block->cb_list); + INIT_LIST_HEAD(&block->owner_list); + INIT_LIST_HEAD(&block->chain0.filter_chain_list); + + block->refcnt = 1; + block->net = net; + block->index = block_index; + + /* Don't store q pointer for blocks which are shared */ + if (!tcf_block_shared(block)) + block->q = q; + return block; +} + +static struct tcf_block *tcf_block_lookup(struct net *net, u32 block_index) +{ + struct tcf_net *tn = net_generic(net, tcf_net_id); + + return idr_find(&tn->idr, block_index); +} + +/* Find tcf block. + * Set q, parent, cl when appropriate. + */ + +static struct tcf_block *tcf_block_find(struct net *net, struct Qdisc **q, + u32 *parent, unsigned long *cl, + int ifindex, u32 block_index, + struct netlink_ext_ack *extack) +{ + struct tcf_block *block; + int err = 0; + + if (ifindex == TCM_IFINDEX_MAGIC_BLOCK) { + block = tcf_block_lookup(net, block_index); + if (!block) { + NL_SET_ERR_MSG(extack, "Block of given index was not found"); + return ERR_PTR(-EINVAL); + } + } else { + const struct Qdisc_class_ops *cops; + struct net_device *dev; + + rcu_read_lock(); + + /* Find link */ + dev = dev_get_by_index_rcu(net, ifindex); + if (!dev) { + rcu_read_unlock(); + return ERR_PTR(-ENODEV); + } + + /* Find qdisc */ + if (!*parent) { + *q = dev->qdisc; + *parent = (*q)->handle; + } else { + *q = qdisc_lookup_rcu(dev, TC_H_MAJ(*parent)); + if (!*q) { + NL_SET_ERR_MSG(extack, "Parent Qdisc doesn't exists"); + err = -EINVAL; + goto errout_rcu; + } + } + + *q = qdisc_refcount_inc_nz(*q); + if (!*q) { + NL_SET_ERR_MSG(extack, "Parent Qdisc doesn't exists"); + err = -EINVAL; + goto errout_rcu; + } + + /* Is it classful? */ + cops = (*q)->ops->cl_ops; + if (!cops) { + NL_SET_ERR_MSG(extack, "Qdisc not classful"); + err = -EINVAL; + goto errout_rcu; + } + + if (!cops->tcf_block) { + NL_SET_ERR_MSG(extack, "Class doesn't support blocks"); + err = -EOPNOTSUPP; + goto errout_rcu; + } + + /* At this point we know that qdisc is not noop_qdisc, + * which means that qdisc holds a reference to net_device + * and we hold a reference to qdisc, so it is safe to release + * rcu read lock. + */ + rcu_read_unlock(); + + /* Do we search for filter, attached to class? */ + if (TC_H_MIN(*parent)) { + *cl = cops->find(*q, *parent); + if (*cl == 0) { + NL_SET_ERR_MSG(extack, "Specified class doesn't exist"); + err = -ENOENT; + goto errout_qdisc; + } + } + + /* And the last stroke */ + block = cops->tcf_block(*q, *cl, extack); + if (!block) { + err = -EINVAL; + goto errout_qdisc; + } + if (tcf_block_shared(block)) { + NL_SET_ERR_MSG(extack, "This filter block is shared. Please use the block index to manipulate the filters"); + err = -EOPNOTSUPP; + goto errout_qdisc; + } + } + + return block; + +errout_rcu: + rcu_read_unlock(); +errout_qdisc: + if (*q) { + qdisc_put(*q); + *q = NULL; + } + return ERR_PTR(err); +} + +static void tcf_block_release(struct Qdisc *q, struct tcf_block *block) +{ + if (q) + qdisc_put(q); +} + +struct tcf_block_owner_item { + struct list_head list; + struct Qdisc *q; + enum tcf_block_binder_type binder_type; +}; + +static void +tcf_block_owner_netif_keep_dst(struct tcf_block *block, + struct Qdisc *q, + enum tcf_block_binder_type binder_type) +{ + if (block->keep_dst && + binder_type != TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS && + binder_type != TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS) + netif_keep_dst(qdisc_dev(q)); +} + +void tcf_block_netif_keep_dst(struct tcf_block *block) +{ + struct tcf_block_owner_item *item; + + block->keep_dst = true; + list_for_each_entry(item, &block->owner_list, list) + tcf_block_owner_netif_keep_dst(block, item->q, + item->binder_type); +} +EXPORT_SYMBOL(tcf_block_netif_keep_dst); + +static int tcf_block_owner_add(struct tcf_block *block, + struct Qdisc *q, + enum tcf_block_binder_type binder_type) +{ + struct tcf_block_owner_item *item; + + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) + return -ENOMEM; + item->q = q; + item->binder_type = binder_type; + list_add(&item->list, &block->owner_list); + return 0; +} + +static void tcf_block_owner_del(struct tcf_block *block, + struct Qdisc *q, + enum tcf_block_binder_type binder_type) +{ + struct tcf_block_owner_item *item; + + list_for_each_entry(item, &block->owner_list, list) { + if (item->q == q && item->binder_type == binder_type) { + list_del(&item->list); + kfree(item); + return; + } + } + WARN_ON(1); +} + +int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q, + struct tcf_block_ext_info *ei, + struct netlink_ext_ack *extack) +{ + struct net *net = qdisc_net(q); + struct tcf_block *block = NULL; + bool created = false; + int err; + + if (ei->block_index) { + /* block_index not 0 means the shared block is requested */ + block = tcf_block_lookup(net, ei->block_index); + if (block) + block->refcnt++; + } + + if (!block) { + block = tcf_block_create(net, q, ei->block_index, extack); + if (IS_ERR(block)) + return PTR_ERR(block); + created = true; + if (tcf_block_shared(block)) { + err = tcf_block_insert(block, net, extack); + if (err) + goto err_block_insert; + } + } + + err = tcf_block_owner_add(block, q, ei->binder_type); + if (err) + goto err_block_owner_add; + + tcf_block_owner_netif_keep_dst(block, q, ei->binder_type); + + err = tcf_chain0_head_change_cb_add(block, ei, extack); + if (err) + goto err_chain0_head_change_cb_add; + + err = tcf_block_offload_bind(block, q, ei, extack); + if (err) + goto err_block_offload_bind; + + *p_block = block; + return 0; + +err_block_offload_bind: + tcf_chain0_head_change_cb_del(block, ei); +err_chain0_head_change_cb_add: + tcf_block_owner_del(block, q, ei->binder_type); +err_block_owner_add: + if (created) { + if (tcf_block_shared(block)) + tcf_block_remove(block, net); +err_block_insert: + kfree(block); + } else { + block->refcnt--; + } + return err; +} +EXPORT_SYMBOL(tcf_block_get_ext); + +static void tcf_chain_head_change_dflt(struct tcf_proto *tp_head, void *priv) +{ + struct tcf_proto __rcu **p_filter_chain = priv; + + rcu_assign_pointer(*p_filter_chain, tp_head); +} + +int tcf_block_get(struct tcf_block **p_block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct netlink_ext_ack *extack) +{ + struct tcf_block_ext_info ei = { + .chain_head_change = tcf_chain_head_change_dflt, + .chain_head_change_priv = p_filter_chain, + }; + + WARN_ON(!p_filter_chain); + return tcf_block_get_ext(p_block, q, &ei, extack); +} +EXPORT_SYMBOL(tcf_block_get); + +/* XXX: Standalone actions are not allowed to jump to any chain, and bound + * actions should be all removed after flushing. + */ +void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q, + struct tcf_block_ext_info *ei) +{ + struct tcf_chain *chain, *tmp; + + if (!block) + return; + tcf_chain0_head_change_cb_del(block, ei); + tcf_block_owner_del(block, q, ei->binder_type); + + if (block->refcnt == 1) { + if (tcf_block_shared(block)) + tcf_block_remove(block, block->net); + + /* Hold a refcnt for all chains, so that they don't disappear + * while we are iterating. + */ + list_for_each_entry(chain, &block->chain_list, list) + tcf_chain_hold(chain); + + list_for_each_entry(chain, &block->chain_list, list) + tcf_chain_flush(chain); + } + + tcf_block_offload_unbind(block, q, ei); + + if (block->refcnt == 1) { + /* At this point, all the chains should have refcnt >= 1. */ + list_for_each_entry_safe(chain, tmp, &block->chain_list, list) { + tcf_chain_put_explicitly_created(chain); + tcf_chain_put(chain); + } + + block->refcnt--; + if (list_empty(&block->chain_list)) + kfree(block); + } else { + block->refcnt--; + } +} +EXPORT_SYMBOL(tcf_block_put_ext); + +void tcf_block_put(struct tcf_block *block) +{ + struct tcf_block_ext_info ei = {0, }; + + if (!block) + return; + tcf_block_put_ext(block, block->q, &ei); +} + +EXPORT_SYMBOL(tcf_block_put); + +struct tcf_block_cb { + struct list_head list; + tc_setup_cb_t *cb; + void *cb_ident; + void *cb_priv; + unsigned int refcnt; +}; + +void *tcf_block_cb_priv(struct tcf_block_cb *block_cb) +{ + return block_cb->cb_priv; +} +EXPORT_SYMBOL(tcf_block_cb_priv); + +struct tcf_block_cb *tcf_block_cb_lookup(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident) +{ struct tcf_block_cb *block_cb; + + list_for_each_entry(block_cb, &block->cb_list, list) + if (block_cb->cb == cb && block_cb->cb_ident == cb_ident) + return block_cb; + return NULL; +} +EXPORT_SYMBOL(tcf_block_cb_lookup); + +void tcf_block_cb_incref(struct tcf_block_cb *block_cb) +{ + block_cb->refcnt++; +} +EXPORT_SYMBOL(tcf_block_cb_incref); + +unsigned int tcf_block_cb_decref(struct tcf_block_cb *block_cb) +{ + return --block_cb->refcnt; +} +EXPORT_SYMBOL(tcf_block_cb_decref); + +static int +tcf_block_playback_offloads(struct tcf_block *block, tc_setup_cb_t *cb, + void *cb_priv, bool add, bool offload_in_use, + struct netlink_ext_ack *extack) +{ + struct tcf_chain *chain; + struct tcf_proto *tp; + int err; + + list_for_each_entry(chain, &block->chain_list, list) { + for (tp = rtnl_dereference(chain->filter_chain); tp; + tp = rtnl_dereference(tp->next)) { + if (tp->ops->reoffload) { + err = tp->ops->reoffload(tp, add, cb, cb_priv, + extack); + if (err && add) + goto err_playback_remove; + } else if (add && offload_in_use) { + err = -EOPNOTSUPP; + NL_SET_ERR_MSG(extack, "Filter HW offload failed - classifier without re-offloading support"); + goto err_playback_remove; + } + } + } + + return 0; + +err_playback_remove: + tcf_block_playback_offloads(block, cb, cb_priv, false, offload_in_use, + extack); + return err; +} + +struct tcf_block_cb *__tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv, + struct netlink_ext_ack *extack) +{ + struct tcf_block_cb *block_cb; + int err; + + /* Replay any already present rules */ + err = tcf_block_playback_offloads(block, cb, cb_priv, true, + tcf_block_offload_in_use(block), + extack); + if (err) + return ERR_PTR(err); + + block_cb = kzalloc(sizeof(*block_cb), GFP_KERNEL); + if (!block_cb) + return ERR_PTR(-ENOMEM); + block_cb->cb = cb; + block_cb->cb_ident = cb_ident; + block_cb->cb_priv = cb_priv; + list_add(&block_cb->list, &block->cb_list); + return block_cb; +} +EXPORT_SYMBOL(__tcf_block_cb_register); + +int tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv, struct netlink_ext_ack *extack) +{ + struct tcf_block_cb *block_cb; + + block_cb = __tcf_block_cb_register(block, cb, cb_ident, cb_priv, + extack); + return PTR_ERR_OR_ZERO(block_cb); +} +EXPORT_SYMBOL(tcf_block_cb_register); + +void __tcf_block_cb_unregister(struct tcf_block *block, + struct tcf_block_cb *block_cb) +{ + tcf_block_playback_offloads(block, block_cb->cb, block_cb->cb_priv, + false, tcf_block_offload_in_use(block), + NULL); + list_del(&block_cb->list); + kfree(block_cb); +} +EXPORT_SYMBOL(__tcf_block_cb_unregister); + +void tcf_block_cb_unregister(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident) +{ + struct tcf_block_cb *block_cb; + + block_cb = tcf_block_cb_lookup(block, cb, cb_ident); + if (!block_cb) + return; + __tcf_block_cb_unregister(block, block_cb); +} +EXPORT_SYMBOL(tcf_block_cb_unregister); + +static int tcf_block_cb_call(struct tcf_block *block, enum tc_setup_type type, + void *type_data, bool err_stop) +{ + struct tcf_block_cb *block_cb; + int ok_count = 0; + int err; + + /* Make sure all netdevs sharing this block are offload-capable. */ + if (block->nooffloaddevcnt && err_stop) + return -EOPNOTSUPP; + + list_for_each_entry(block_cb, &block->cb_list, list) { + err = block_cb->cb(type, type_data, block_cb->cb_priv); + if (err) { + if (err_stop) + return err; + } else { + ok_count++; + } + } + return ok_count; +} + +/* Main classifier routine: scans classifier chain attached + * to this qdisc, (optionally) tests for protocol and asks + * specific classifiers. + */ +int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res, bool compat_mode) +{ +#ifdef CONFIG_NET_CLS_ACT + const int max_reclassify_loop = 4; + const struct tcf_proto *orig_tp = tp; + const struct tcf_proto *first_tp; + int limit = 0; + +reclassify: +#endif + for (; tp; tp = rcu_dereference_bh(tp->next)) { + __be16 protocol = skb_protocol(skb, false); + int err; + + if (tp->protocol != protocol && + tp->protocol != htons(ETH_P_ALL)) + continue; + + err = tp->classify(skb, tp, res); +#ifdef CONFIG_NET_CLS_ACT + if (unlikely(err == TC_ACT_RECLASSIFY && !compat_mode)) { + first_tp = orig_tp; + goto reset; + } else if (unlikely(TC_ACT_EXT_CMP(err, TC_ACT_GOTO_CHAIN))) { + first_tp = res->goto_tp; + goto reset; + } +#endif + if (err >= 0) + return err; + } + + return TC_ACT_UNSPEC; /* signal: continue lookup */ +#ifdef CONFIG_NET_CLS_ACT +reset: + if (unlikely(limit++ >= max_reclassify_loop)) { + net_notice_ratelimited("%u: reclassify loop, rule prio %u, protocol %02x\n", + tp->chain->block->index, + tp->prio & 0xffff, + ntohs(tp->protocol)); + return TC_ACT_SHOT; + } + + tp = first_tp; + goto reclassify; +#endif +} +EXPORT_SYMBOL(tcf_classify); + +struct tcf_chain_info { + struct tcf_proto __rcu **pprev; + struct tcf_proto __rcu *next; +}; + +static struct tcf_proto *tcf_chain_tp_prev(struct tcf_chain_info *chain_info) +{ + return rtnl_dereference(*chain_info->pprev); +} + +static void tcf_chain_tp_insert(struct tcf_chain *chain, + struct tcf_chain_info *chain_info, + struct tcf_proto *tp) +{ + if (*chain_info->pprev == chain->filter_chain) + tcf_chain0_head_change(chain, tp); + RCU_INIT_POINTER(tp->next, tcf_chain_tp_prev(chain_info)); + rcu_assign_pointer(*chain_info->pprev, tp); + tcf_chain_hold(chain); +} + +static void tcf_chain_tp_remove(struct tcf_chain *chain, + struct tcf_chain_info *chain_info, + struct tcf_proto *tp) +{ + struct tcf_proto *next = rtnl_dereference(chain_info->next); + + if (tp == chain->filter_chain) + tcf_chain0_head_change(chain, next); + RCU_INIT_POINTER(*chain_info->pprev, next); + tcf_chain_put(chain); +} + +static struct tcf_proto *tcf_chain_tp_find(struct tcf_chain *chain, + struct tcf_chain_info *chain_info, + u32 protocol, u32 prio, + bool prio_allocate) +{ + struct tcf_proto **pprev; + struct tcf_proto *tp; + + /* Check the chain for existence of proto-tcf with this priority */ + for (pprev = &chain->filter_chain; + (tp = rtnl_dereference(*pprev)); pprev = &tp->next) { + if (tp->prio >= prio) { + if (tp->prio == prio) { + if (prio_allocate || + (tp->protocol != protocol && protocol)) + return ERR_PTR(-EINVAL); + } else { + tp = NULL; + } + break; + } + } + chain_info->pprev = pprev; + chain_info->next = tp ? tp->next : NULL; + return tp; +} + +static int tcf_fill_node(struct net *net, struct sk_buff *skb, + struct tcf_proto *tp, struct tcf_block *block, + struct Qdisc *q, u32 parent, void *fh, + u32 portid, u32 seq, u16 flags, int event) +{ + struct tcmsg *tcm; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + if (q) { + tcm->tcm_ifindex = qdisc_dev(q)->ifindex; + tcm->tcm_parent = parent; + } else { + tcm->tcm_ifindex = TCM_IFINDEX_MAGIC_BLOCK; + tcm->tcm_block_index = block->index; + } + tcm->tcm_info = TC_H_MAKE(tp->prio, tp->protocol); + if (nla_put_string(skb, TCA_KIND, tp->ops->kind)) + goto nla_put_failure; + if (nla_put_u32(skb, TCA_CHAIN, tp->chain->index)) + goto nla_put_failure; + if (!fh) { + tcm->tcm_handle = 0; + } else { + if (tp->ops->dump && tp->ops->dump(net, tp, fh, skb, tcm) < 0) + goto nla_put_failure; + } + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int tfilter_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, struct tcf_proto *tp, + struct tcf_block *block, struct Qdisc *q, + u32 parent, void *fh, int event, bool unicast) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tcf_fill_node(net, skb, tp, block, q, parent, fh, portid, + n->nlmsg_seq, n->nlmsg_flags, event) <= 0) { + kfree_skb(skb); + return -EINVAL; + } + + if (unicast) + return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT); + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +static int tfilter_del_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, struct tcf_proto *tp, + struct tcf_block *block, struct Qdisc *q, + u32 parent, void *fh, bool unicast, bool *last, + struct netlink_ext_ack *extack) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + int err; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tcf_fill_node(net, skb, tp, block, q, parent, fh, portid, + n->nlmsg_seq, n->nlmsg_flags, RTM_DELTFILTER) <= 0) { + NL_SET_ERR_MSG(extack, "Failed to build del event notification"); + kfree_skb(skb); + return -EINVAL; + } + + err = tp->ops->delete(tp, fh, last, extack); + if (err) { + kfree_skb(skb); + return err; + } + + if (unicast) + return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT); + + err = rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); + if (err < 0) + NL_SET_ERR_MSG(extack, "Failed to send filter delete notification"); + return err; +} + +static void tfilter_notify_chain(struct net *net, struct sk_buff *oskb, + struct tcf_block *block, struct Qdisc *q, + u32 parent, struct nlmsghdr *n, + struct tcf_chain *chain, int event) +{ + struct tcf_proto *tp; + + for (tp = rtnl_dereference(chain->filter_chain); + tp; tp = rtnl_dereference(tp->next)) + tfilter_notify(net, oskb, n, tp, block, + q, parent, NULL, event, false); +} + +static int tc_new_tfilter(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_MAX + 1]; + struct tcmsg *t; + u32 protocol; + u32 prio; + bool prio_allocate; + u32 parent; + u32 chain_index; + struct Qdisc *q = NULL; + struct tcf_chain_info chain_info; + struct tcf_chain *chain = NULL; + struct tcf_block *block; + struct tcf_proto *tp; + unsigned long cl; + void *fh; + int err; + int tp_created; + + if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + +replay: + tp_created = 0; + + err = nlmsg_parse(n, sizeof(*t), tca, TCA_MAX, rtm_tca_policy, extack); + if (err < 0) + return err; + + t = nlmsg_data(n); + protocol = TC_H_MIN(t->tcm_info); + prio = TC_H_MAJ(t->tcm_info); + prio_allocate = false; + parent = t->tcm_parent; + cl = 0; + + if (prio == 0) { + /* If no priority is provided by the user, + * we allocate one. + */ + if (n->nlmsg_flags & NLM_F_CREATE) { + prio = TC_H_MAKE(0x80000000U, 0U); + prio_allocate = true; + } else { + NL_SET_ERR_MSG(extack, "Invalid filter command with priority of zero"); + return -ENOENT; + } + } + + /* Find head of filter chain. */ + + block = tcf_block_find(net, &q, &parent, &cl, + t->tcm_ifindex, t->tcm_block_index, extack); + if (IS_ERR(block)) { + err = PTR_ERR(block); + goto errout; + } + + chain_index = tca[TCA_CHAIN] ? nla_get_u32(tca[TCA_CHAIN]) : 0; + if (chain_index > TC_ACT_EXT_VAL_MASK) { + NL_SET_ERR_MSG(extack, "Specified chain index exceeds upper limit"); + err = -EINVAL; + goto errout; + } + chain = tcf_chain_get(block, chain_index, true); + if (!chain) { + NL_SET_ERR_MSG(extack, "Cannot create specified filter chain"); + err = -ENOMEM; + goto errout; + } + + tp = tcf_chain_tp_find(chain, &chain_info, protocol, + prio, prio_allocate); + if (IS_ERR(tp)) { + NL_SET_ERR_MSG(extack, "Filter with specified priority/protocol not found"); + err = PTR_ERR(tp); + goto errout; + } + + if (tp == NULL) { + /* Proto-tcf does not exist, create new one */ + + if (tca[TCA_KIND] == NULL || !protocol) { + NL_SET_ERR_MSG(extack, "Filter kind and protocol must be specified"); + err = -EINVAL; + goto errout; + } + + if (!(n->nlmsg_flags & NLM_F_CREATE)) { + NL_SET_ERR_MSG(extack, "Need both RTM_NEWTFILTER and NLM_F_CREATE to create a new filter"); + err = -ENOENT; + goto errout; + } + + if (prio_allocate) + prio = tcf_auto_prio(tcf_chain_tp_prev(&chain_info)); + + tp = tcf_proto_create(nla_data(tca[TCA_KIND]), + protocol, prio, chain, extack); + if (IS_ERR(tp)) { + err = PTR_ERR(tp); + goto errout; + } + tp_created = 1; + } else if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], tp->ops->kind)) { + NL_SET_ERR_MSG(extack, "Specified filter kind does not match existing one"); + err = -EINVAL; + goto errout; + } + + fh = tp->ops->get(tp, t->tcm_handle); + + if (!fh) { + if (!(n->nlmsg_flags & NLM_F_CREATE)) { + NL_SET_ERR_MSG(extack, "Need both RTM_NEWTFILTER and NLM_F_CREATE to create a new filter"); + err = -ENOENT; + goto errout; + } + } else if (n->nlmsg_flags & NLM_F_EXCL) { + NL_SET_ERR_MSG(extack, "Filter already exists"); + err = -EEXIST; + goto errout; + } + + if (chain->tmplt_ops && chain->tmplt_ops != tp->ops) { + NL_SET_ERR_MSG(extack, "Chain template is set to a different filter kind"); + err = -EINVAL; + goto errout; + } + + err = tp->ops->change(net, skb, tp, cl, t->tcm_handle, tca, &fh, + n->nlmsg_flags & NLM_F_CREATE ? TCA_ACT_NOREPLACE : TCA_ACT_REPLACE, + extack); + if (err == 0) { + if (tp_created) + tcf_chain_tp_insert(chain, &chain_info, tp); + tfilter_notify(net, skb, n, tp, block, q, parent, fh, + RTM_NEWTFILTER, false); + /* q pointer is NULL for shared blocks */ + if (q) + q->flags &= ~TCQ_F_CAN_BYPASS; + } else { + if (tp_created) + tcf_proto_destroy(tp, NULL); + } + +errout: + if (chain) + tcf_chain_put(chain); + tcf_block_release(q, block); + if (err == -EAGAIN) + /* Replay the request. */ + goto replay; + return err; +} + +static int tc_del_tfilter(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_MAX + 1]; + struct tcmsg *t; + u32 protocol; + u32 prio; + u32 parent; + u32 chain_index; + struct Qdisc *q = NULL; + struct tcf_chain_info chain_info; + struct tcf_chain *chain = NULL; + struct tcf_block *block; + struct tcf_proto *tp = NULL; + unsigned long cl = 0; + void *fh = NULL; + int err; + + if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + err = nlmsg_parse(n, sizeof(*t), tca, TCA_MAX, rtm_tca_policy, extack); + if (err < 0) + return err; + + t = nlmsg_data(n); + protocol = TC_H_MIN(t->tcm_info); + prio = TC_H_MAJ(t->tcm_info); + parent = t->tcm_parent; + + if (prio == 0 && (protocol || t->tcm_handle || tca[TCA_KIND])) { + NL_SET_ERR_MSG(extack, "Cannot flush filters with protocol, handle or kind set"); + return -ENOENT; + } + + /* Find head of filter chain. */ + + block = tcf_block_find(net, &q, &parent, &cl, + t->tcm_ifindex, t->tcm_block_index, extack); + if (IS_ERR(block)) { + err = PTR_ERR(block); + goto errout; + } + + chain_index = tca[TCA_CHAIN] ? nla_get_u32(tca[TCA_CHAIN]) : 0; + if (chain_index > TC_ACT_EXT_VAL_MASK) { + NL_SET_ERR_MSG(extack, "Specified chain index exceeds upper limit"); + err = -EINVAL; + goto errout; + } + chain = tcf_chain_get(block, chain_index, false); + if (!chain) { + /* User requested flush on non-existent chain. Nothing to do, + * so just return success. + */ + if (prio == 0) { + err = 0; + goto errout; + } + NL_SET_ERR_MSG(extack, "Cannot find specified filter chain"); + err = -ENOENT; + goto errout; + } + + if (prio == 0) { + tfilter_notify_chain(net, skb, block, q, parent, n, + chain, RTM_DELTFILTER); + tcf_chain_flush(chain); + err = 0; + goto errout; + } + + tp = tcf_chain_tp_find(chain, &chain_info, protocol, + prio, false); + if (!tp || IS_ERR(tp)) { + NL_SET_ERR_MSG(extack, "Filter with specified priority/protocol not found"); + err = tp ? PTR_ERR(tp) : -ENOENT; + goto errout; + } else if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], tp->ops->kind)) { + NL_SET_ERR_MSG(extack, "Specified filter kind does not match existing one"); + err = -EINVAL; + goto errout; + } + + fh = tp->ops->get(tp, t->tcm_handle); + + if (!fh) { + if (t->tcm_handle == 0) { + tcf_chain_tp_remove(chain, &chain_info, tp); + tfilter_notify(net, skb, n, tp, block, q, parent, fh, + RTM_DELTFILTER, false); + tcf_proto_destroy(tp, extack); + err = 0; + } else { + NL_SET_ERR_MSG(extack, "Specified filter handle not found"); + err = -ENOENT; + } + } else { + bool last; + + err = tfilter_del_notify(net, skb, n, tp, block, + q, parent, fh, false, &last, + extack); + if (err) + goto errout; + if (last) { + tcf_chain_tp_remove(chain, &chain_info, tp); + tcf_proto_destroy(tp, extack); + } + } + +errout: + if (chain) + tcf_chain_put(chain); + tcf_block_release(q, block); + return err; +} + +static int tc_get_tfilter(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_MAX + 1]; + struct tcmsg *t; + u32 protocol; + u32 prio; + u32 parent; + u32 chain_index; + struct Qdisc *q = NULL; + struct tcf_chain_info chain_info; + struct tcf_chain *chain = NULL; + struct tcf_block *block; + struct tcf_proto *tp = NULL; + unsigned long cl = 0; + void *fh = NULL; + int err; + + err = nlmsg_parse(n, sizeof(*t), tca, TCA_MAX, rtm_tca_policy, extack); + if (err < 0) + return err; + + t = nlmsg_data(n); + protocol = TC_H_MIN(t->tcm_info); + prio = TC_H_MAJ(t->tcm_info); + parent = t->tcm_parent; + + if (prio == 0) { + NL_SET_ERR_MSG(extack, "Invalid filter command with priority of zero"); + return -ENOENT; + } + + /* Find head of filter chain. */ + + block = tcf_block_find(net, &q, &parent, &cl, + t->tcm_ifindex, t->tcm_block_index, extack); + if (IS_ERR(block)) { + err = PTR_ERR(block); + goto errout; + } + + chain_index = tca[TCA_CHAIN] ? nla_get_u32(tca[TCA_CHAIN]) : 0; + if (chain_index > TC_ACT_EXT_VAL_MASK) { + NL_SET_ERR_MSG(extack, "Specified chain index exceeds upper limit"); + err = -EINVAL; + goto errout; + } + chain = tcf_chain_get(block, chain_index, false); + if (!chain) { + NL_SET_ERR_MSG(extack, "Cannot find specified filter chain"); + err = -EINVAL; + goto errout; + } + + tp = tcf_chain_tp_find(chain, &chain_info, protocol, + prio, false); + if (!tp || IS_ERR(tp)) { + NL_SET_ERR_MSG(extack, "Filter with specified priority/protocol not found"); + err = tp ? PTR_ERR(tp) : -ENOENT; + goto errout; + } else if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], tp->ops->kind)) { + NL_SET_ERR_MSG(extack, "Specified filter kind does not match existing one"); + err = -EINVAL; + goto errout; + } + + fh = tp->ops->get(tp, t->tcm_handle); + + if (!fh) { + NL_SET_ERR_MSG(extack, "Specified filter handle not found"); + err = -ENOENT; + } else { + err = tfilter_notify(net, skb, n, tp, block, q, parent, + fh, RTM_NEWTFILTER, true); + if (err < 0) + NL_SET_ERR_MSG(extack, "Failed to send filter notify message"); + } + +errout: + if (chain) + tcf_chain_put(chain); + tcf_block_release(q, block); + return err; +} + +struct tcf_dump_args { + struct tcf_walker w; + struct sk_buff *skb; + struct netlink_callback *cb; + struct tcf_block *block; + struct Qdisc *q; + u32 parent; +}; + +static int tcf_node_dump(struct tcf_proto *tp, void *n, struct tcf_walker *arg) +{ + struct tcf_dump_args *a = (void *)arg; + struct net *net = sock_net(a->skb->sk); + + return tcf_fill_node(net, a->skb, tp, a->block, a->q, a->parent, + n, NETLINK_CB(a->cb->skb).portid, + a->cb->nlh->nlmsg_seq, NLM_F_MULTI, + RTM_NEWTFILTER); +} + +static bool tcf_chain_dump(struct tcf_chain *chain, struct Qdisc *q, u32 parent, + struct sk_buff *skb, struct netlink_callback *cb, + long index_start, long *p_index) +{ + struct net *net = sock_net(skb->sk); + struct tcf_block *block = chain->block; + struct tcmsg *tcm = nlmsg_data(cb->nlh); + struct tcf_dump_args arg; + struct tcf_proto *tp; + + for (tp = rtnl_dereference(chain->filter_chain); + tp; tp = rtnl_dereference(tp->next), (*p_index)++) { + if (*p_index < index_start) + continue; + if (TC_H_MAJ(tcm->tcm_info) && + TC_H_MAJ(tcm->tcm_info) != tp->prio) + continue; + if (TC_H_MIN(tcm->tcm_info) && + TC_H_MIN(tcm->tcm_info) != tp->protocol) + continue; + if (*p_index > index_start) + memset(&cb->args[1], 0, + sizeof(cb->args) - sizeof(cb->args[0])); + if (cb->args[1] == 0) { + if (tcf_fill_node(net, skb, tp, block, q, parent, NULL, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + RTM_NEWTFILTER) <= 0) + return false; + + cb->args[1] = 1; + } + if (!tp->ops->walk) + continue; + arg.w.fn = tcf_node_dump; + arg.skb = skb; + arg.cb = cb; + arg.block = block; + arg.q = q; + arg.parent = parent; + arg.w.stop = 0; + arg.w.skip = cb->args[1] - 1; + arg.w.count = 0; + arg.w.cookie = cb->args[2]; + tp->ops->walk(tp, &arg.w); + cb->args[2] = arg.w.cookie; + cb->args[1] = arg.w.count + 1; + if (arg.w.stop) + return false; + } + return true; +} + +/* called with RTNL */ +static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_MAX + 1]; + struct Qdisc *q = NULL; + struct tcf_block *block; + struct tcf_chain *chain; + struct tcmsg *tcm = nlmsg_data(cb->nlh); + long index_start; + long index; + u32 parent; + int err; + + if (nlmsg_len(cb->nlh) < sizeof(*tcm)) + return skb->len; + + err = nlmsg_parse(cb->nlh, sizeof(*tcm), tca, TCA_MAX, NULL, NULL); + if (err) + return err; + + if (tcm->tcm_ifindex == TCM_IFINDEX_MAGIC_BLOCK) { + block = tcf_block_lookup(net, tcm->tcm_block_index); + if (!block) + goto out; + /* If we work with block index, q is NULL and parent value + * will never be used in the following code. The check + * in tcf_fill_node prevents it. However, compiler does not + * see that far, so set parent to zero to silence the warning + * about parent being uninitialized. + */ + parent = 0; + } else { + const struct Qdisc_class_ops *cops; + struct net_device *dev; + unsigned long cl = 0; + + dev = __dev_get_by_index(net, tcm->tcm_ifindex); + if (!dev) + return skb->len; + + parent = tcm->tcm_parent; + if (!parent) { + q = dev->qdisc; + parent = q->handle; + } else { + q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent)); + } + if (!q) + goto out; + cops = q->ops->cl_ops; + if (!cops) + goto out; + if (!cops->tcf_block) + goto out; + if (TC_H_MIN(tcm->tcm_parent)) { + cl = cops->find(q, tcm->tcm_parent); + if (cl == 0) + goto out; + } + block = cops->tcf_block(q, cl, NULL); + if (!block) + goto out; + if (tcf_block_shared(block)) + q = NULL; + } + + index_start = cb->args[0]; + index = 0; + + list_for_each_entry(chain, &block->chain_list, list) { + if (tca[TCA_CHAIN] && + nla_get_u32(tca[TCA_CHAIN]) != chain->index) + continue; + if (!tcf_chain_dump(chain, q, parent, skb, cb, + index_start, &index)) { + err = -EMSGSIZE; + break; + } + } + + cb->args[0] = index; + +out: + /* If we did no progress, the error (EMSGSIZE) is real */ + if (skb->len == 0 && err) + return err; + return skb->len; +} + +static int tc_chain_fill_node(struct tcf_chain *chain, struct net *net, + struct sk_buff *skb, struct tcf_block *block, + u32 portid, u32 seq, u16 flags, int event) +{ + unsigned char *b = skb_tail_pointer(skb); + const struct tcf_proto_ops *ops; + struct nlmsghdr *nlh; + struct tcmsg *tcm; + void *priv; + + ops = chain->tmplt_ops; + priv = chain->tmplt_priv; + + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + tcm->tcm_handle = 0; + if (block->q) { + tcm->tcm_ifindex = qdisc_dev(block->q)->ifindex; + tcm->tcm_parent = block->q->handle; + } else { + tcm->tcm_ifindex = TCM_IFINDEX_MAGIC_BLOCK; + tcm->tcm_block_index = block->index; + } + + if (nla_put_u32(skb, TCA_CHAIN, chain->index)) + goto nla_put_failure; + + if (ops) { + if (nla_put_string(skb, TCA_KIND, ops->kind)) + goto nla_put_failure; + if (ops->tmplt_dump(skb, net, priv) < 0) + goto nla_put_failure; + } + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -EMSGSIZE; +} + +static int tc_chain_notify(struct tcf_chain *chain, struct sk_buff *oskb, + u32 seq, u16 flags, int event, bool unicast) +{ + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + struct tcf_block *block = chain->block; + struct net *net = block->net; + struct sk_buff *skb; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tc_chain_fill_node(chain, net, skb, block, portid, + seq, flags, event) <= 0) { + kfree_skb(skb); + return -EINVAL; + } + + if (unicast) + return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT); + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, flags & NLM_F_ECHO); +} + +static int tc_chain_tmplt_add(struct tcf_chain *chain, struct net *net, + struct nlattr **tca, + struct netlink_ext_ack *extack) +{ + const struct tcf_proto_ops *ops; + void *tmplt_priv; + + /* If kind is not set, user did not specify template. */ + if (!tca[TCA_KIND]) + return 0; + + ops = tcf_proto_lookup_ops(nla_data(tca[TCA_KIND]), extack); + if (IS_ERR(ops)) + return PTR_ERR(ops); + if (!ops->tmplt_create || !ops->tmplt_destroy || !ops->tmplt_dump) { + NL_SET_ERR_MSG(extack, "Chain templates are not supported with specified classifier"); + return -EOPNOTSUPP; + } + + tmplt_priv = ops->tmplt_create(net, chain, tca, extack); + if (IS_ERR(tmplt_priv)) { + module_put(ops->owner); + return PTR_ERR(tmplt_priv); + } + chain->tmplt_ops = ops; + chain->tmplt_priv = tmplt_priv; + return 0; +} + +static void tc_chain_tmplt_del(struct tcf_chain *chain) +{ + const struct tcf_proto_ops *ops = chain->tmplt_ops; + + /* If template ops are set, no work to do for us. */ + if (!ops) + return; + + ops->tmplt_destroy(chain->tmplt_priv); + module_put(ops->owner); +} + +/* Add/delete/get a chain */ + +static int tc_ctl_chain(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_MAX + 1]; + struct tcmsg *t; + u32 parent; + u32 chain_index; + struct Qdisc *q = NULL; + struct tcf_chain *chain = NULL; + struct tcf_block *block; + unsigned long cl; + int err; + + if (n->nlmsg_type != RTM_GETCHAIN && + !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + +replay: + err = nlmsg_parse(n, sizeof(*t), tca, TCA_MAX, rtm_tca_policy, extack); + if (err < 0) + return err; + + t = nlmsg_data(n); + parent = t->tcm_parent; + cl = 0; + + block = tcf_block_find(net, &q, &parent, &cl, + t->tcm_ifindex, t->tcm_block_index, extack); + if (IS_ERR(block)) + return PTR_ERR(block); + + chain_index = tca[TCA_CHAIN] ? nla_get_u32(tca[TCA_CHAIN]) : 0; + if (chain_index > TC_ACT_EXT_VAL_MASK) { + NL_SET_ERR_MSG(extack, "Specified chain index exceeds upper limit"); + err = -EINVAL; + goto errout_block; + } + chain = tcf_chain_lookup(block, chain_index); + if (n->nlmsg_type == RTM_NEWCHAIN) { + if (chain) { + if (tcf_chain_held_by_acts_only(chain)) { + /* The chain exists only because there is + * some action referencing it. + */ + tcf_chain_hold(chain); + } else { + NL_SET_ERR_MSG(extack, "Filter chain already exists"); + err = -EEXIST; + goto errout_block; + } + } else { + if (!(n->nlmsg_flags & NLM_F_CREATE)) { + NL_SET_ERR_MSG(extack, "Need both RTM_NEWCHAIN and NLM_F_CREATE to create a new chain"); + err = -ENOENT; + goto errout_block; + } + chain = tcf_chain_create(block, chain_index); + if (!chain) { + NL_SET_ERR_MSG(extack, "Failed to create filter chain"); + err = -ENOMEM; + goto errout_block; + } + } + } else { + if (!chain || tcf_chain_held_by_acts_only(chain)) { + NL_SET_ERR_MSG(extack, "Cannot find specified filter chain"); + err = -EINVAL; + goto errout_block; + } + tcf_chain_hold(chain); + } + + switch (n->nlmsg_type) { + case RTM_NEWCHAIN: + err = tc_chain_tmplt_add(chain, net, tca, extack); + if (err) + goto errout; + /* In case the chain was successfully added, take a reference + * to the chain. This ensures that an empty chain + * does not disappear at the end of this function. + */ + tcf_chain_hold(chain); + chain->explicitly_created = true; + tc_chain_notify(chain, NULL, 0, NLM_F_CREATE | NLM_F_EXCL, + RTM_NEWCHAIN, false); + break; + case RTM_DELCHAIN: + tfilter_notify_chain(net, skb, block, q, parent, n, + chain, RTM_DELTFILTER); + /* Flush the chain first as the user requested chain removal. */ + tcf_chain_flush(chain); + /* In case the chain was successfully deleted, put a reference + * to the chain previously taken during addition. + */ + tcf_chain_put_explicitly_created(chain); + chain->explicitly_created = false; + break; + case RTM_GETCHAIN: + err = tc_chain_notify(chain, skb, n->nlmsg_seq, + n->nlmsg_flags, n->nlmsg_type, true); + if (err < 0) + NL_SET_ERR_MSG(extack, "Failed to send chain notify message"); + break; + default: + err = -EOPNOTSUPP; + NL_SET_ERR_MSG(extack, "Unsupported message type"); + goto errout; + } + +errout: + tcf_chain_put(chain); +errout_block: + tcf_block_release(q, block); + if (err == -EAGAIN) + /* Replay the request. */ + goto replay; + return err; +} + +/* called with RTNL */ +static int tc_dump_chain(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + struct nlattr *tca[TCA_MAX + 1]; + struct Qdisc *q = NULL; + struct tcf_block *block; + struct tcf_chain *chain; + struct tcmsg *tcm = nlmsg_data(cb->nlh); + long index_start; + long index; + u32 parent; + int err; + + if (nlmsg_len(cb->nlh) < sizeof(*tcm)) + return skb->len; + + err = nlmsg_parse(cb->nlh, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, + NULL); + if (err) + return err; + + if (tcm->tcm_ifindex == TCM_IFINDEX_MAGIC_BLOCK) { + block = tcf_block_lookup(net, tcm->tcm_block_index); + if (!block) + goto out; + /* If we work with block index, q is NULL and parent value + * will never be used in the following code. The check + * in tcf_fill_node prevents it. However, compiler does not + * see that far, so set parent to zero to silence the warning + * about parent being uninitialized. + */ + parent = 0; + } else { + const struct Qdisc_class_ops *cops; + struct net_device *dev; + unsigned long cl = 0; + + dev = __dev_get_by_index(net, tcm->tcm_ifindex); + if (!dev) + return skb->len; + + parent = tcm->tcm_parent; + if (!parent) { + q = dev->qdisc; + parent = q->handle; + } else { + q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent)); + } + if (!q) + goto out; + cops = q->ops->cl_ops; + if (!cops) + goto out; + if (!cops->tcf_block) + goto out; + if (TC_H_MIN(tcm->tcm_parent)) { + cl = cops->find(q, tcm->tcm_parent); + if (cl == 0) + goto out; + } + block = cops->tcf_block(q, cl, NULL); + if (!block) + goto out; + if (tcf_block_shared(block)) + q = NULL; + } + + index_start = cb->args[0]; + index = 0; + + list_for_each_entry(chain, &block->chain_list, list) { + if ((tca[TCA_CHAIN] && + nla_get_u32(tca[TCA_CHAIN]) != chain->index)) + continue; + if (index < index_start) { + index++; + continue; + } + if (tcf_chain_held_by_acts_only(chain)) + continue; + err = tc_chain_fill_node(chain, net, skb, block, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + RTM_NEWCHAIN); + if (err <= 0) + break; + index++; + } + + cb->args[0] = index; + +out: + /* If we did no progress, the error (EMSGSIZE) is real */ + if (skb->len == 0 && err) + return err; + return skb->len; +} + +void tcf_exts_destroy(struct tcf_exts *exts) +{ +#ifdef CONFIG_NET_CLS_ACT + if (exts->actions) { + tcf_action_destroy(exts->actions, TCA_ACT_UNBIND); + kfree(exts->actions); + } + exts->nr_actions = 0; +#endif +} +EXPORT_SYMBOL(tcf_exts_destroy); + +int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb, + struct nlattr *rate_tlv, struct tcf_exts *exts, bool ovr, + struct netlink_ext_ack *extack) +{ +#ifdef CONFIG_NET_CLS_ACT + { + struct tc_action *act; + size_t attr_size = 0; + + if (exts->police && tb[exts->police]) { + act = tcf_action_init_1(net, tp, tb[exts->police], + rate_tlv, "police", ovr, + TCA_ACT_BIND, true, extack); + if (IS_ERR(act)) + return PTR_ERR(act); + + act->type = exts->type = TCA_OLD_COMPAT; + exts->actions[0] = act; + exts->nr_actions = 1; + } else if (exts->action && tb[exts->action]) { + int err; + + err = tcf_action_init(net, tp, tb[exts->action], + rate_tlv, NULL, ovr, TCA_ACT_BIND, + exts->actions, &attr_size, true, + extack); + if (err < 0) + return err; + exts->nr_actions = err; + } + exts->net = net; + } +#else + if ((exts->action && tb[exts->action]) || + (exts->police && tb[exts->police])) { + NL_SET_ERR_MSG(extack, "Classifier actions are not supported per compile options (CONFIG_NET_CLS_ACT)"); + return -EOPNOTSUPP; + } +#endif + + return 0; +} +EXPORT_SYMBOL(tcf_exts_validate); + +void tcf_exts_change(struct tcf_exts *dst, struct tcf_exts *src) +{ +#ifdef CONFIG_NET_CLS_ACT + struct tcf_exts old = *dst; + + *dst = *src; + tcf_exts_destroy(&old); +#endif +} +EXPORT_SYMBOL(tcf_exts_change); + +#ifdef CONFIG_NET_CLS_ACT +static struct tc_action *tcf_exts_first_act(struct tcf_exts *exts) +{ + if (exts->nr_actions == 0) + return NULL; + else + return exts->actions[0]; +} +#endif + +int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts) +{ +#ifdef CONFIG_NET_CLS_ACT + struct nlattr *nest; + + if (exts->action && tcf_exts_has_actions(exts)) { + /* + * again for backward compatible mode - we want + * to work with both old and new modes of entering + * tc data even if iproute2 was newer - jhs + */ + if (exts->type != TCA_OLD_COMPAT) { + nest = nla_nest_start(skb, exts->action); + if (nest == NULL) + goto nla_put_failure; + + if (tcf_action_dump(skb, exts->actions, 0, 0) < 0) + goto nla_put_failure; + nla_nest_end(skb, nest); + } else if (exts->police) { + struct tc_action *act = tcf_exts_first_act(exts); + nest = nla_nest_start(skb, exts->police); + if (nest == NULL || !act) + goto nla_put_failure; + if (tcf_action_dump_old(skb, act, 0, 0) < 0) + goto nla_put_failure; + nla_nest_end(skb, nest); + } + } + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +#else + return 0; +#endif +} +EXPORT_SYMBOL(tcf_exts_dump); + + +int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts) +{ +#ifdef CONFIG_NET_CLS_ACT + struct tc_action *a = tcf_exts_first_act(exts); + if (a != NULL && tcf_action_copy_stats(skb, a, 1) < 0) + return -1; +#endif + return 0; +} +EXPORT_SYMBOL(tcf_exts_dump_stats); + +static int tc_exts_setup_cb_egdev_call(struct tcf_exts *exts, + enum tc_setup_type type, + void *type_data, bool err_stop) +{ + int ok_count = 0; +#ifdef CONFIG_NET_CLS_ACT + const struct tc_action *a; + struct net_device *dev; + int i, ret; + + if (!tcf_exts_has_actions(exts)) + return 0; + + for (i = 0; i < exts->nr_actions; i++) { + a = exts->actions[i]; + if (!a->ops->get_dev) + continue; + dev = a->ops->get_dev(a); + if (!dev) + continue; + ret = tc_setup_cb_egdev_call(dev, type, type_data, err_stop); + a->ops->put_dev(dev); + if (ret < 0) + return ret; + ok_count += ret; + } +#endif + return ok_count; +} + +int tc_setup_cb_call(struct tcf_block *block, struct tcf_exts *exts, + enum tc_setup_type type, void *type_data, bool err_stop) +{ + int ok_count; + int ret; + + ret = tcf_block_cb_call(block, type, type_data, err_stop); + if (ret < 0) + return ret; + ok_count = ret; + + if (!exts || ok_count) + return ok_count; + ret = tc_exts_setup_cb_egdev_call(exts, type, type_data, err_stop); + if (ret < 0) + return ret; + ok_count += ret; + + return ok_count; +} +EXPORT_SYMBOL(tc_setup_cb_call); + +static __net_init int tcf_net_init(struct net *net) +{ + struct tcf_net *tn = net_generic(net, tcf_net_id); + + idr_init(&tn->idr); + return 0; +} + +static void __net_exit tcf_net_exit(struct net *net) +{ + struct tcf_net *tn = net_generic(net, tcf_net_id); + + idr_destroy(&tn->idr); +} + +static struct pernet_operations tcf_net_ops = { + .init = tcf_net_init, + .exit = tcf_net_exit, + .id = &tcf_net_id, + .size = sizeof(struct tcf_net), +}; + +static int __init tc_filter_init(void) +{ + int err; + + tc_filter_wq = alloc_ordered_workqueue("tc_filter_workqueue", 0); + if (!tc_filter_wq) + return -ENOMEM; + + err = register_pernet_subsys(&tcf_net_ops); + if (err) + goto err_register_pernet_subsys; + + rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_new_tfilter, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELTFILTER, tc_del_tfilter, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETTFILTER, tc_get_tfilter, + tc_dump_tfilter, 0); + rtnl_register(PF_UNSPEC, RTM_NEWCHAIN, tc_ctl_chain, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELCHAIN, tc_ctl_chain, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETCHAIN, tc_ctl_chain, + tc_dump_chain, 0); + + return 0; + +err_register_pernet_subsys: + destroy_workqueue(tc_filter_wq); + return err; +} + +subsys_initcall(tc_filter_init); diff --git a/net/sched/cls_basic.c b/net/sched/cls_basic.c new file mode 100644 index 000000000..14098da69 --- /dev/null +++ b/net/sched/cls_basic.c @@ -0,0 +1,331 @@ +/* + * net/sched/cls_basic.c Basic Packet Classifier. + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/rtnetlink.h> +#include <linux/skbuff.h> +#include <linux/idr.h> +#include <net/netlink.h> +#include <net/act_api.h> +#include <net/pkt_cls.h> + +struct basic_head { + struct list_head flist; + struct idr handle_idr; + struct rcu_head rcu; +}; + +struct basic_filter { + u32 handle; + struct tcf_exts exts; + struct tcf_ematch_tree ematches; + struct tcf_result res; + struct tcf_proto *tp; + struct list_head link; + struct rcu_work rwork; +}; + +static int basic_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + int r; + struct basic_head *head = rcu_dereference_bh(tp->root); + struct basic_filter *f; + + list_for_each_entry_rcu(f, &head->flist, link) { + if (!tcf_em_tree_match(skb, &f->ematches, NULL)) + continue; + *res = f->res; + r = tcf_exts_exec(skb, &f->exts, res); + if (r < 0) + continue; + return r; + } + return -1; +} + +static void *basic_get(struct tcf_proto *tp, u32 handle) +{ + struct basic_head *head = rtnl_dereference(tp->root); + struct basic_filter *f; + + list_for_each_entry(f, &head->flist, link) { + if (f->handle == handle) { + return f; + } + } + + return NULL; +} + +static int basic_init(struct tcf_proto *tp) +{ + struct basic_head *head; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (head == NULL) + return -ENOBUFS; + INIT_LIST_HEAD(&head->flist); + idr_init(&head->handle_idr); + rcu_assign_pointer(tp->root, head); + return 0; +} + +static void __basic_delete_filter(struct basic_filter *f) +{ + tcf_exts_destroy(&f->exts); + tcf_em_tree_destroy(&f->ematches); + tcf_exts_put_net(&f->exts); + kfree(f); +} + +static void basic_delete_filter_work(struct work_struct *work) +{ + struct basic_filter *f = container_of(to_rcu_work(work), + struct basic_filter, + rwork); + rtnl_lock(); + __basic_delete_filter(f); + rtnl_unlock(); +} + +static void basic_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct basic_head *head = rtnl_dereference(tp->root); + struct basic_filter *f, *n; + + list_for_each_entry_safe(f, n, &head->flist, link) { + list_del_rcu(&f->link); + tcf_unbind_filter(tp, &f->res); + idr_remove(&head->handle_idr, f->handle); + if (tcf_exts_get_net(&f->exts)) + tcf_queue_work(&f->rwork, basic_delete_filter_work); + else + __basic_delete_filter(f); + } + idr_destroy(&head->handle_idr); + kfree_rcu(head, rcu); +} + +static int basic_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct basic_head *head = rtnl_dereference(tp->root); + struct basic_filter *f = arg; + + list_del_rcu(&f->link); + tcf_unbind_filter(tp, &f->res); + idr_remove(&head->handle_idr, f->handle); + tcf_exts_get_net(&f->exts); + tcf_queue_work(&f->rwork, basic_delete_filter_work); + *last = list_empty(&head->flist); + return 0; +} + +static const struct nla_policy basic_policy[TCA_BASIC_MAX + 1] = { + [TCA_BASIC_CLASSID] = { .type = NLA_U32 }, + [TCA_BASIC_EMATCHES] = { .type = NLA_NESTED }, +}; + +static int basic_set_parms(struct net *net, struct tcf_proto *tp, + struct basic_filter *f, unsigned long base, + struct nlattr **tb, + struct nlattr *est, bool ovr, + struct netlink_ext_ack *extack) +{ + int err; + + err = tcf_exts_validate(net, tp, tb, est, &f->exts, ovr, extack); + if (err < 0) + return err; + + err = tcf_em_tree_validate(tp, tb[TCA_BASIC_EMATCHES], &f->ematches); + if (err < 0) + return err; + + if (tb[TCA_BASIC_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_BASIC_CLASSID]); + tcf_bind_filter(tp, &f->res, base); + } + + f->tp = tp; + return 0; +} + +static int basic_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, u32 handle, + struct nlattr **tca, void **arg, bool ovr, + struct netlink_ext_ack *extack) +{ + int err; + struct basic_head *head = rtnl_dereference(tp->root); + struct nlattr *tb[TCA_BASIC_MAX + 1]; + struct basic_filter *fold = (struct basic_filter *) *arg; + struct basic_filter *fnew; + + if (tca[TCA_OPTIONS] == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_BASIC_MAX, tca[TCA_OPTIONS], + basic_policy, NULL); + if (err < 0) + return err; + + if (fold != NULL) { + if (handle && fold->handle != handle) + return -EINVAL; + } + + fnew = kzalloc(sizeof(*fnew), GFP_KERNEL); + if (!fnew) + return -ENOBUFS; + + err = tcf_exts_init(&fnew->exts, TCA_BASIC_ACT, TCA_BASIC_POLICE); + if (err < 0) + goto errout; + + if (!handle) { + handle = 1; + err = idr_alloc_u32(&head->handle_idr, fnew, &handle, + INT_MAX, GFP_KERNEL); + } else if (!fold) { + err = idr_alloc_u32(&head->handle_idr, fnew, &handle, + handle, GFP_KERNEL); + } + if (err) + goto errout; + fnew->handle = handle; + + err = basic_set_parms(net, tp, fnew, base, tb, tca[TCA_RATE], ovr, + extack); + if (err < 0) { + if (!fold) + idr_remove(&head->handle_idr, fnew->handle); + goto errout; + } + + *arg = fnew; + + if (fold) { + idr_replace(&head->handle_idr, fnew, fnew->handle); + list_replace_rcu(&fold->link, &fnew->link); + tcf_unbind_filter(tp, &fold->res); + tcf_exts_get_net(&fold->exts); + tcf_queue_work(&fold->rwork, basic_delete_filter_work); + } else { + list_add_rcu(&fnew->link, &head->flist); + } + + return 0; +errout: + tcf_exts_destroy(&fnew->exts); + kfree(fnew); + return err; +} + +static void basic_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct basic_head *head = rtnl_dereference(tp->root); + struct basic_filter *f; + + list_for_each_entry(f, &head->flist, link) { + if (arg->count < arg->skip) + goto skip; + + if (arg->fn(tp, f, arg) < 0) { + arg->stop = 1; + break; + } +skip: + arg->count++; + } +} + +static void basic_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct basic_filter *f = fh; + + if (f && f->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &f->res, base); + else + __tcf_unbind_filter(q, &f->res); + } +} + +static int basic_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct basic_filter *f = fh; + struct nlattr *nest; + + if (f == NULL) + return skb->len; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (f->res.classid && + nla_put_u32(skb, TCA_BASIC_CLASSID, f->res.classid)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts) < 0 || + tcf_em_tree_dump(skb, &f->ematches, TCA_BASIC_EMATCHES) < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static struct tcf_proto_ops cls_basic_ops __read_mostly = { + .kind = "basic", + .classify = basic_classify, + .init = basic_init, + .destroy = basic_destroy, + .get = basic_get, + .change = basic_change, + .delete = basic_delete, + .walk = basic_walk, + .dump = basic_dump, + .bind_class = basic_bind_class, + .owner = THIS_MODULE, +}; + +static int __init init_basic(void) +{ + return register_tcf_proto_ops(&cls_basic_ops); +} + +static void __exit exit_basic(void) +{ + unregister_tcf_proto_ops(&cls_basic_ops); +} + +module_init(init_basic) +module_exit(exit_basic) +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c new file mode 100644 index 000000000..5d100126c --- /dev/null +++ b/net/sched/cls_bpf.c @@ -0,0 +1,722 @@ +/* + * Berkeley Packet Filter based traffic classifier + * + * Might be used to classify traffic through flexible, user-defined and + * possibly JIT-ed BPF filters for traffic control as an alternative to + * ematches. + * + * (C) 2013 Daniel Borkmann <dborkman@redhat.com> + * + * 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/types.h> +#include <linux/skbuff.h> +#include <linux/filter.h> +#include <linux/bpf.h> +#include <linux/idr.h> + +#include <net/rtnetlink.h> +#include <net/pkt_cls.h> +#include <net/sock.h> + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Daniel Borkmann <dborkman@redhat.com>"); +MODULE_DESCRIPTION("TC BPF based classifier"); + +#define CLS_BPF_NAME_LEN 256 +#define CLS_BPF_SUPPORTED_GEN_FLAGS \ + (TCA_CLS_FLAGS_SKIP_HW | TCA_CLS_FLAGS_SKIP_SW) + +struct cls_bpf_head { + struct list_head plist; + struct idr handle_idr; + struct rcu_head rcu; +}; + +struct cls_bpf_prog { + struct bpf_prog *filter; + struct list_head link; + struct tcf_result res; + bool exts_integrated; + u32 gen_flags; + unsigned int in_hw_count; + struct tcf_exts exts; + u32 handle; + u16 bpf_num_ops; + struct sock_filter *bpf_ops; + const char *bpf_name; + struct tcf_proto *tp; + struct rcu_work rwork; +}; + +static const struct nla_policy bpf_policy[TCA_BPF_MAX + 1] = { + [TCA_BPF_CLASSID] = { .type = NLA_U32 }, + [TCA_BPF_FLAGS] = { .type = NLA_U32 }, + [TCA_BPF_FLAGS_GEN] = { .type = NLA_U32 }, + [TCA_BPF_FD] = { .type = NLA_U32 }, + [TCA_BPF_NAME] = { .type = NLA_NUL_STRING, + .len = CLS_BPF_NAME_LEN }, + [TCA_BPF_OPS_LEN] = { .type = NLA_U16 }, + [TCA_BPF_OPS] = { .type = NLA_BINARY, + .len = sizeof(struct sock_filter) * BPF_MAXINSNS }, +}; + +static int cls_bpf_exec_opcode(int code) +{ + switch (code) { + case TC_ACT_OK: + case TC_ACT_SHOT: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + case TC_ACT_REDIRECT: + case TC_ACT_UNSPEC: + return code; + default: + return TC_ACT_UNSPEC; + } +} + +static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct cls_bpf_head *head = rcu_dereference_bh(tp->root); + bool at_ingress = skb_at_tc_ingress(skb); + struct cls_bpf_prog *prog; + int ret = -1; + + /* Needed here for accessing maps. */ + rcu_read_lock(); + list_for_each_entry_rcu(prog, &head->plist, link) { + int filter_res; + + qdisc_skb_cb(skb)->tc_classid = prog->res.classid; + + if (tc_skip_sw(prog->gen_flags)) { + filter_res = prog->exts_integrated ? TC_ACT_UNSPEC : 0; + } else if (at_ingress) { + /* It is safe to push/pull even if skb_shared() */ + __skb_push(skb, skb->mac_len); + bpf_compute_data_pointers(skb); + filter_res = BPF_PROG_RUN(prog->filter, skb); + __skb_pull(skb, skb->mac_len); + } else { + bpf_compute_data_pointers(skb); + filter_res = BPF_PROG_RUN(prog->filter, skb); + } + + if (prog->exts_integrated) { + res->class = 0; + res->classid = TC_H_MAJ(prog->res.classid) | + qdisc_skb_cb(skb)->tc_classid; + + ret = cls_bpf_exec_opcode(filter_res); + if (ret == TC_ACT_UNSPEC) + continue; + break; + } + + if (filter_res == 0) + continue; + if (filter_res != -1) { + res->class = 0; + res->classid = filter_res; + } else { + *res = prog->res; + } + + ret = tcf_exts_exec(skb, &prog->exts, res); + if (ret < 0) + continue; + + break; + } + rcu_read_unlock(); + + return ret; +} + +static bool cls_bpf_is_ebpf(const struct cls_bpf_prog *prog) +{ + return !prog->bpf_ops; +} + +static int cls_bpf_offload_cmd(struct tcf_proto *tp, struct cls_bpf_prog *prog, + struct cls_bpf_prog *oldprog, + struct netlink_ext_ack *extack) +{ + struct tcf_block *block = tp->chain->block; + struct tc_cls_bpf_offload cls_bpf = {}; + struct cls_bpf_prog *obj; + bool skip_sw; + int err; + + skip_sw = prog && tc_skip_sw(prog->gen_flags); + obj = prog ?: oldprog; + + tc_cls_common_offload_init(&cls_bpf.common, tp, obj->gen_flags, + extack); + cls_bpf.command = TC_CLSBPF_OFFLOAD; + cls_bpf.exts = &obj->exts; + cls_bpf.prog = prog ? prog->filter : NULL; + cls_bpf.oldprog = oldprog ? oldprog->filter : NULL; + cls_bpf.name = obj->bpf_name; + cls_bpf.exts_integrated = obj->exts_integrated; + + if (oldprog) + tcf_block_offload_dec(block, &oldprog->gen_flags); + + err = tc_setup_cb_call(block, NULL, TC_SETUP_CLSBPF, &cls_bpf, skip_sw); + if (prog) { + if (err < 0) { + cls_bpf_offload_cmd(tp, oldprog, prog, extack); + return err; + } else if (err > 0) { + prog->in_hw_count = err; + tcf_block_offload_inc(block, &prog->gen_flags); + } + } + + if (prog && skip_sw && !(prog->gen_flags & TCA_CLS_FLAGS_IN_HW)) + return -EINVAL; + + return 0; +} + +static u32 cls_bpf_flags(u32 flags) +{ + return flags & CLS_BPF_SUPPORTED_GEN_FLAGS; +} + +static int cls_bpf_offload(struct tcf_proto *tp, struct cls_bpf_prog *prog, + struct cls_bpf_prog *oldprog, + struct netlink_ext_ack *extack) +{ + if (prog && oldprog && + cls_bpf_flags(prog->gen_flags) != + cls_bpf_flags(oldprog->gen_flags)) + return -EINVAL; + + if (prog && tc_skip_hw(prog->gen_flags)) + prog = NULL; + if (oldprog && tc_skip_hw(oldprog->gen_flags)) + oldprog = NULL; + if (!prog && !oldprog) + return 0; + + return cls_bpf_offload_cmd(tp, prog, oldprog, extack); +} + +static void cls_bpf_stop_offload(struct tcf_proto *tp, + struct cls_bpf_prog *prog, + struct netlink_ext_ack *extack) +{ + int err; + + err = cls_bpf_offload_cmd(tp, NULL, prog, extack); + if (err) + pr_err("Stopping hardware offload failed: %d\n", err); +} + +static void cls_bpf_offload_update_stats(struct tcf_proto *tp, + struct cls_bpf_prog *prog) +{ + struct tcf_block *block = tp->chain->block; + struct tc_cls_bpf_offload cls_bpf = {}; + + tc_cls_common_offload_init(&cls_bpf.common, tp, prog->gen_flags, NULL); + cls_bpf.command = TC_CLSBPF_STATS; + cls_bpf.exts = &prog->exts; + cls_bpf.prog = prog->filter; + cls_bpf.name = prog->bpf_name; + cls_bpf.exts_integrated = prog->exts_integrated; + + tc_setup_cb_call(block, NULL, TC_SETUP_CLSBPF, &cls_bpf, false); +} + +static int cls_bpf_init(struct tcf_proto *tp) +{ + struct cls_bpf_head *head; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (head == NULL) + return -ENOBUFS; + + INIT_LIST_HEAD_RCU(&head->plist); + idr_init(&head->handle_idr); + rcu_assign_pointer(tp->root, head); + + return 0; +} + +static void cls_bpf_free_parms(struct cls_bpf_prog *prog) +{ + if (cls_bpf_is_ebpf(prog)) + bpf_prog_put(prog->filter); + else + bpf_prog_destroy(prog->filter); + + kfree(prog->bpf_name); + kfree(prog->bpf_ops); +} + +static void __cls_bpf_delete_prog(struct cls_bpf_prog *prog) +{ + tcf_exts_destroy(&prog->exts); + tcf_exts_put_net(&prog->exts); + + cls_bpf_free_parms(prog); + kfree(prog); +} + +static void cls_bpf_delete_prog_work(struct work_struct *work) +{ + struct cls_bpf_prog *prog = container_of(to_rcu_work(work), + struct cls_bpf_prog, + rwork); + rtnl_lock(); + __cls_bpf_delete_prog(prog); + rtnl_unlock(); +} + +static void __cls_bpf_delete(struct tcf_proto *tp, struct cls_bpf_prog *prog, + struct netlink_ext_ack *extack) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + + idr_remove(&head->handle_idr, prog->handle); + cls_bpf_stop_offload(tp, prog, extack); + list_del_rcu(&prog->link); + tcf_unbind_filter(tp, &prog->res); + if (tcf_exts_get_net(&prog->exts)) + tcf_queue_work(&prog->rwork, cls_bpf_delete_prog_work); + else + __cls_bpf_delete_prog(prog); +} + +static int cls_bpf_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + + __cls_bpf_delete(tp, arg, extack); + *last = list_empty(&head->plist); + return 0; +} + +static void cls_bpf_destroy(struct tcf_proto *tp, + struct netlink_ext_ack *extack) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + struct cls_bpf_prog *prog, *tmp; + + list_for_each_entry_safe(prog, tmp, &head->plist, link) + __cls_bpf_delete(tp, prog, extack); + + idr_destroy(&head->handle_idr); + kfree_rcu(head, rcu); +} + +static void *cls_bpf_get(struct tcf_proto *tp, u32 handle) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + struct cls_bpf_prog *prog; + + list_for_each_entry(prog, &head->plist, link) { + if (prog->handle == handle) + return prog; + } + + return NULL; +} + +static int cls_bpf_prog_from_ops(struct nlattr **tb, struct cls_bpf_prog *prog) +{ + struct sock_filter *bpf_ops; + struct sock_fprog_kern fprog_tmp; + struct bpf_prog *fp; + u16 bpf_size, bpf_num_ops; + int ret; + + bpf_num_ops = nla_get_u16(tb[TCA_BPF_OPS_LEN]); + if (bpf_num_ops > BPF_MAXINSNS || bpf_num_ops == 0) + return -EINVAL; + + bpf_size = bpf_num_ops * sizeof(*bpf_ops); + if (bpf_size != nla_len(tb[TCA_BPF_OPS])) + return -EINVAL; + + bpf_ops = kmemdup(nla_data(tb[TCA_BPF_OPS]), bpf_size, GFP_KERNEL); + if (bpf_ops == NULL) + return -ENOMEM; + + fprog_tmp.len = bpf_num_ops; + fprog_tmp.filter = bpf_ops; + + ret = bpf_prog_create(&fp, &fprog_tmp); + if (ret < 0) { + kfree(bpf_ops); + return ret; + } + + prog->bpf_ops = bpf_ops; + prog->bpf_num_ops = bpf_num_ops; + prog->bpf_name = NULL; + prog->filter = fp; + + return 0; +} + +static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog, + u32 gen_flags, const struct tcf_proto *tp) +{ + struct bpf_prog *fp; + char *name = NULL; + bool skip_sw; + u32 bpf_fd; + + bpf_fd = nla_get_u32(tb[TCA_BPF_FD]); + skip_sw = gen_flags & TCA_CLS_FLAGS_SKIP_SW; + + fp = bpf_prog_get_type_dev(bpf_fd, BPF_PROG_TYPE_SCHED_CLS, skip_sw); + if (IS_ERR(fp)) + return PTR_ERR(fp); + + if (tb[TCA_BPF_NAME]) { + name = nla_memdup(tb[TCA_BPF_NAME], GFP_KERNEL); + if (!name) { + bpf_prog_put(fp); + return -ENOMEM; + } + } + + prog->bpf_ops = NULL; + prog->bpf_name = name; + prog->filter = fp; + + if (fp->dst_needed) + tcf_block_netif_keep_dst(tp->chain->block); + + return 0; +} + +static int cls_bpf_set_parms(struct net *net, struct tcf_proto *tp, + struct cls_bpf_prog *prog, unsigned long base, + struct nlattr **tb, struct nlattr *est, bool ovr, + struct netlink_ext_ack *extack) +{ + bool is_bpf, is_ebpf, have_exts = false; + u32 gen_flags = 0; + int ret; + + is_bpf = tb[TCA_BPF_OPS_LEN] && tb[TCA_BPF_OPS]; + is_ebpf = tb[TCA_BPF_FD]; + if ((!is_bpf && !is_ebpf) || (is_bpf && is_ebpf)) + return -EINVAL; + + ret = tcf_exts_validate(net, tp, tb, est, &prog->exts, ovr, extack); + if (ret < 0) + return ret; + + if (tb[TCA_BPF_FLAGS]) { + u32 bpf_flags = nla_get_u32(tb[TCA_BPF_FLAGS]); + + if (bpf_flags & ~TCA_BPF_FLAG_ACT_DIRECT) + return -EINVAL; + + have_exts = bpf_flags & TCA_BPF_FLAG_ACT_DIRECT; + } + if (tb[TCA_BPF_FLAGS_GEN]) { + gen_flags = nla_get_u32(tb[TCA_BPF_FLAGS_GEN]); + if (gen_flags & ~CLS_BPF_SUPPORTED_GEN_FLAGS || + !tc_flags_valid(gen_flags)) + return -EINVAL; + } + + prog->exts_integrated = have_exts; + prog->gen_flags = gen_flags; + + ret = is_bpf ? cls_bpf_prog_from_ops(tb, prog) : + cls_bpf_prog_from_efd(tb, prog, gen_flags, tp); + if (ret < 0) + return ret; + + if (tb[TCA_BPF_CLASSID]) { + prog->res.classid = nla_get_u32(tb[TCA_BPF_CLASSID]); + tcf_bind_filter(tp, &prog->res, base); + } + + return 0; +} + +static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + void **arg, bool ovr, struct netlink_ext_ack *extack) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + struct cls_bpf_prog *oldprog = *arg; + struct nlattr *tb[TCA_BPF_MAX + 1]; + struct cls_bpf_prog *prog; + int ret; + + if (tca[TCA_OPTIONS] == NULL) + return -EINVAL; + + ret = nla_parse_nested(tb, TCA_BPF_MAX, tca[TCA_OPTIONS], bpf_policy, + NULL); + if (ret < 0) + return ret; + + prog = kzalloc(sizeof(*prog), GFP_KERNEL); + if (!prog) + return -ENOBUFS; + + ret = tcf_exts_init(&prog->exts, TCA_BPF_ACT, TCA_BPF_POLICE); + if (ret < 0) + goto errout; + + if (oldprog) { + if (handle && oldprog->handle != handle) { + ret = -EINVAL; + goto errout; + } + } + + if (handle == 0) { + handle = 1; + ret = idr_alloc_u32(&head->handle_idr, prog, &handle, + INT_MAX, GFP_KERNEL); + } else if (!oldprog) { + ret = idr_alloc_u32(&head->handle_idr, prog, &handle, + handle, GFP_KERNEL); + } + + if (ret) + goto errout; + prog->handle = handle; + + ret = cls_bpf_set_parms(net, tp, prog, base, tb, tca[TCA_RATE], ovr, + extack); + if (ret < 0) + goto errout_idr; + + ret = cls_bpf_offload(tp, prog, oldprog, extack); + if (ret) + goto errout_parms; + + if (!tc_in_hw(prog->gen_flags)) + prog->gen_flags |= TCA_CLS_FLAGS_NOT_IN_HW; + + if (oldprog) { + idr_replace(&head->handle_idr, prog, handle); + list_replace_rcu(&oldprog->link, &prog->link); + tcf_unbind_filter(tp, &oldprog->res); + tcf_exts_get_net(&oldprog->exts); + tcf_queue_work(&oldprog->rwork, cls_bpf_delete_prog_work); + } else { + list_add_rcu(&prog->link, &head->plist); + } + + *arg = prog; + return 0; + +errout_parms: + cls_bpf_free_parms(prog); +errout_idr: + if (!oldprog) + idr_remove(&head->handle_idr, prog->handle); +errout: + tcf_exts_destroy(&prog->exts); + kfree(prog); + return ret; +} + +static int cls_bpf_dump_bpf_info(const struct cls_bpf_prog *prog, + struct sk_buff *skb) +{ + struct nlattr *nla; + + if (nla_put_u16(skb, TCA_BPF_OPS_LEN, prog->bpf_num_ops)) + return -EMSGSIZE; + + nla = nla_reserve(skb, TCA_BPF_OPS, prog->bpf_num_ops * + sizeof(struct sock_filter)); + if (nla == NULL) + return -EMSGSIZE; + + memcpy(nla_data(nla), prog->bpf_ops, nla_len(nla)); + + return 0; +} + +static int cls_bpf_dump_ebpf_info(const struct cls_bpf_prog *prog, + struct sk_buff *skb) +{ + struct nlattr *nla; + + if (prog->bpf_name && + nla_put_string(skb, TCA_BPF_NAME, prog->bpf_name)) + return -EMSGSIZE; + + if (nla_put_u32(skb, TCA_BPF_ID, prog->filter->aux->id)) + return -EMSGSIZE; + + nla = nla_reserve(skb, TCA_BPF_TAG, sizeof(prog->filter->tag)); + if (nla == NULL) + return -EMSGSIZE; + + memcpy(nla_data(nla), prog->filter->tag, nla_len(nla)); + + return 0; +} + +static int cls_bpf_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *tm) +{ + struct cls_bpf_prog *prog = fh; + struct nlattr *nest; + u32 bpf_flags = 0; + int ret; + + if (prog == NULL) + return skb->len; + + tm->tcm_handle = prog->handle; + + cls_bpf_offload_update_stats(tp, prog); + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (prog->res.classid && + nla_put_u32(skb, TCA_BPF_CLASSID, prog->res.classid)) + goto nla_put_failure; + + if (cls_bpf_is_ebpf(prog)) + ret = cls_bpf_dump_ebpf_info(prog, skb); + else + ret = cls_bpf_dump_bpf_info(prog, skb); + if (ret) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &prog->exts) < 0) + goto nla_put_failure; + + if (prog->exts_integrated) + bpf_flags |= TCA_BPF_FLAG_ACT_DIRECT; + if (bpf_flags && nla_put_u32(skb, TCA_BPF_FLAGS, bpf_flags)) + goto nla_put_failure; + if (prog->gen_flags && + nla_put_u32(skb, TCA_BPF_FLAGS_GEN, prog->gen_flags)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &prog->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void cls_bpf_bind_class(void *fh, u32 classid, unsigned long cl, + void *q, unsigned long base) +{ + struct cls_bpf_prog *prog = fh; + + if (prog && prog->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &prog->res, base); + else + __tcf_unbind_filter(q, &prog->res); + } +} + +static void cls_bpf_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + struct cls_bpf_prog *prog; + + list_for_each_entry(prog, &head->plist, link) { + if (arg->count < arg->skip) + goto skip; + if (arg->fn(tp, prog, arg) < 0) { + arg->stop = 1; + break; + } +skip: + arg->count++; + } +} + +static int cls_bpf_reoffload(struct tcf_proto *tp, bool add, tc_setup_cb_t *cb, + void *cb_priv, struct netlink_ext_ack *extack) +{ + struct cls_bpf_head *head = rtnl_dereference(tp->root); + struct tcf_block *block = tp->chain->block; + struct tc_cls_bpf_offload cls_bpf = {}; + struct cls_bpf_prog *prog; + int err; + + list_for_each_entry(prog, &head->plist, link) { + if (tc_skip_hw(prog->gen_flags)) + continue; + + tc_cls_common_offload_init(&cls_bpf.common, tp, prog->gen_flags, + extack); + cls_bpf.command = TC_CLSBPF_OFFLOAD; + cls_bpf.exts = &prog->exts; + cls_bpf.prog = add ? prog->filter : NULL; + cls_bpf.oldprog = add ? NULL : prog->filter; + cls_bpf.name = prog->bpf_name; + cls_bpf.exts_integrated = prog->exts_integrated; + + err = cb(TC_SETUP_CLSBPF, &cls_bpf, cb_priv); + if (err) { + if (add && tc_skip_sw(prog->gen_flags)) + return err; + continue; + } + + tc_cls_offload_cnt_update(block, &prog->in_hw_count, + &prog->gen_flags, add); + } + + return 0; +} + +static struct tcf_proto_ops cls_bpf_ops __read_mostly = { + .kind = "bpf", + .owner = THIS_MODULE, + .classify = cls_bpf_classify, + .init = cls_bpf_init, + .destroy = cls_bpf_destroy, + .get = cls_bpf_get, + .change = cls_bpf_change, + .delete = cls_bpf_delete, + .walk = cls_bpf_walk, + .reoffload = cls_bpf_reoffload, + .dump = cls_bpf_dump, + .bind_class = cls_bpf_bind_class, +}; + +static int __init cls_bpf_init_mod(void) +{ + return register_tcf_proto_ops(&cls_bpf_ops); +} + +static void __exit cls_bpf_exit_mod(void) +{ + unregister_tcf_proto_ops(&cls_bpf_ops); +} + +module_init(cls_bpf_init_mod); +module_exit(cls_bpf_exit_mod); diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c new file mode 100644 index 000000000..3bc01bdde --- /dev/null +++ b/net/sched/cls_cgroup.c @@ -0,0 +1,221 @@ +/* + * net/sched/cls_cgroup.c Control Group Classifier + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/skbuff.h> +#include <linux/rcupdate.h> +#include <net/rtnetlink.h> +#include <net/pkt_cls.h> +#include <net/sock.h> +#include <net/cls_cgroup.h> + +struct cls_cgroup_head { + u32 handle; + struct tcf_exts exts; + struct tcf_ematch_tree ematches; + struct tcf_proto *tp; + struct rcu_work rwork; +}; + +static int cls_cgroup_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct cls_cgroup_head *head = rcu_dereference_bh(tp->root); + u32 classid = task_get_classid(skb); + + if (!classid) + return -1; + if (!tcf_em_tree_match(skb, &head->ematches, NULL)) + return -1; + + res->classid = classid; + res->class = 0; + + return tcf_exts_exec(skb, &head->exts, res); +} + +static void *cls_cgroup_get(struct tcf_proto *tp, u32 handle) +{ + return NULL; +} + +static int cls_cgroup_init(struct tcf_proto *tp) +{ + return 0; +} + +static const struct nla_policy cgroup_policy[TCA_CGROUP_MAX + 1] = { + [TCA_CGROUP_EMATCHES] = { .type = NLA_NESTED }, +}; + +static void __cls_cgroup_destroy(struct cls_cgroup_head *head) +{ + tcf_exts_destroy(&head->exts); + tcf_em_tree_destroy(&head->ematches); + tcf_exts_put_net(&head->exts); + kfree(head); +} + +static void cls_cgroup_destroy_work(struct work_struct *work) +{ + struct cls_cgroup_head *head = container_of(to_rcu_work(work), + struct cls_cgroup_head, + rwork); + rtnl_lock(); + __cls_cgroup_destroy(head); + rtnl_unlock(); +} + +static int cls_cgroup_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + void **arg, bool ovr, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_CGROUP_MAX + 1]; + struct cls_cgroup_head *head = rtnl_dereference(tp->root); + struct cls_cgroup_head *new; + int err; + + if (!tca[TCA_OPTIONS]) + return -EINVAL; + + if (!head && !handle) + return -EINVAL; + + if (head && handle != head->handle) + return -ENOENT; + + new = kzalloc(sizeof(*head), GFP_KERNEL); + if (!new) + return -ENOBUFS; + + err = tcf_exts_init(&new->exts, TCA_CGROUP_ACT, TCA_CGROUP_POLICE); + if (err < 0) + goto errout; + new->handle = handle; + new->tp = tp; + err = nla_parse_nested(tb, TCA_CGROUP_MAX, tca[TCA_OPTIONS], + cgroup_policy, NULL); + if (err < 0) + goto errout; + + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &new->exts, ovr, + extack); + if (err < 0) + goto errout; + + err = tcf_em_tree_validate(tp, tb[TCA_CGROUP_EMATCHES], &new->ematches); + if (err < 0) + goto errout; + + rcu_assign_pointer(tp->root, new); + if (head) { + tcf_exts_get_net(&head->exts); + tcf_queue_work(&head->rwork, cls_cgroup_destroy_work); + } + return 0; +errout: + tcf_exts_destroy(&new->exts); + kfree(new); + return err; +} + +static void cls_cgroup_destroy(struct tcf_proto *tp, + struct netlink_ext_ack *extack) +{ + struct cls_cgroup_head *head = rtnl_dereference(tp->root); + + /* Head can still be NULL due to cls_cgroup_init(). */ + if (head) { + if (tcf_exts_get_net(&head->exts)) + tcf_queue_work(&head->rwork, cls_cgroup_destroy_work); + else + __cls_cgroup_destroy(head); + } +} + +static int cls_cgroup_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + return -EOPNOTSUPP; +} + +static void cls_cgroup_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct cls_cgroup_head *head = rtnl_dereference(tp->root); + + if (arg->count < arg->skip) + goto skip; + + if (arg->fn(tp, head, arg) < 0) { + arg->stop = 1; + return; + } +skip: + arg->count++; +} + +static int cls_cgroup_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct cls_cgroup_head *head = rtnl_dereference(tp->root); + struct nlattr *nest; + + t->tcm_handle = head->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &head->exts) < 0 || + tcf_em_tree_dump(skb, &head->ematches, TCA_CGROUP_EMATCHES) < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &head->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static struct tcf_proto_ops cls_cgroup_ops __read_mostly = { + .kind = "cgroup", + .init = cls_cgroup_init, + .change = cls_cgroup_change, + .classify = cls_cgroup_classify, + .destroy = cls_cgroup_destroy, + .get = cls_cgroup_get, + .delete = cls_cgroup_delete, + .walk = cls_cgroup_walk, + .dump = cls_cgroup_dump, + .owner = THIS_MODULE, +}; + +static int __init init_cgroup_cls(void) +{ + return register_tcf_proto_ops(&cls_cgroup_ops); +} + +static void __exit exit_cgroup_cls(void) +{ + unregister_tcf_proto_ops(&cls_cgroup_ops); +} + +module_init(init_cgroup_cls); +module_exit(exit_cgroup_cls); +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c new file mode 100644 index 000000000..55bf75cb1 --- /dev/null +++ b/net/sched/cls_flow.c @@ -0,0 +1,725 @@ +/* + * net/sched/cls_flow.c Generic flow classifier + * + * Copyright (c) 2007, 2008 Patrick McHardy <kaber@trash.net> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/list.h> +#include <linux/jhash.h> +#include <linux/random.h> +#include <linux/pkt_cls.h> +#include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/if_vlan.h> +#include <linux/slab.h> +#include <linux/module.h> +#include <net/inet_sock.h> + +#include <net/pkt_cls.h> +#include <net/ip.h> +#include <net/route.h> +#include <net/flow_dissector.h> + +#if IS_ENABLED(CONFIG_NF_CONNTRACK) +#include <net/netfilter/nf_conntrack.h> +#endif + +struct flow_head { + struct list_head filters; + struct rcu_head rcu; +}; + +struct flow_filter { + struct list_head list; + struct tcf_exts exts; + struct tcf_ematch_tree ematches; + struct tcf_proto *tp; + struct timer_list perturb_timer; + u32 perturb_period; + u32 handle; + + u32 nkeys; + u32 keymask; + u32 mode; + u32 mask; + u32 xor; + u32 rshift; + u32 addend; + u32 divisor; + u32 baseclass; + u32 hashrnd; + struct rcu_work rwork; +}; + +static inline u32 addr_fold(void *addr) +{ + unsigned long a = (unsigned long)addr; + + return (a & 0xFFFFFFFF) ^ (BITS_PER_LONG > 32 ? a >> 32 : 0); +} + +static u32 flow_get_src(const struct sk_buff *skb, const struct flow_keys *flow) +{ + __be32 src = flow_get_u32_src(flow); + + if (src) + return ntohl(src); + + return addr_fold(skb->sk); +} + +static u32 flow_get_dst(const struct sk_buff *skb, const struct flow_keys *flow) +{ + __be32 dst = flow_get_u32_dst(flow); + + if (dst) + return ntohl(dst); + + return addr_fold(skb_dst(skb)) ^ (__force u16)skb_protocol(skb, true); +} + +static u32 flow_get_proto(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + return flow->basic.ip_proto; +} + +static u32 flow_get_proto_src(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + if (flow->ports.ports) + return ntohs(flow->ports.src); + + return addr_fold(skb->sk); +} + +static u32 flow_get_proto_dst(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + if (flow->ports.ports) + return ntohs(flow->ports.dst); + + return addr_fold(skb_dst(skb)) ^ (__force u16)skb_protocol(skb, true); +} + +static u32 flow_get_iif(const struct sk_buff *skb) +{ + return skb->skb_iif; +} + +static u32 flow_get_priority(const struct sk_buff *skb) +{ + return skb->priority; +} + +static u32 flow_get_mark(const struct sk_buff *skb) +{ + return skb->mark; +} + +static u32 flow_get_nfct(const struct sk_buff *skb) +{ +#if IS_ENABLED(CONFIG_NF_CONNTRACK) + return addr_fold(skb_nfct(skb)); +#else + return 0; +#endif +} + +#if IS_ENABLED(CONFIG_NF_CONNTRACK) +#define CTTUPLE(skb, member) \ +({ \ + enum ip_conntrack_info ctinfo; \ + const struct nf_conn *ct = nf_ct_get(skb, &ctinfo); \ + if (ct == NULL) \ + goto fallback; \ + ct->tuplehash[CTINFO2DIR(ctinfo)].tuple.member; \ +}) +#else +#define CTTUPLE(skb, member) \ +({ \ + goto fallback; \ + 0; \ +}) +#endif + +static u32 flow_get_nfct_src(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + return ntohl(CTTUPLE(skb, src.u3.ip)); + case htons(ETH_P_IPV6): + return ntohl(CTTUPLE(skb, src.u3.ip6[3])); + } +fallback: + return flow_get_src(skb, flow); +} + +static u32 flow_get_nfct_dst(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + return ntohl(CTTUPLE(skb, dst.u3.ip)); + case htons(ETH_P_IPV6): + return ntohl(CTTUPLE(skb, dst.u3.ip6[3])); + } +fallback: + return flow_get_dst(skb, flow); +} + +static u32 flow_get_nfct_proto_src(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + return ntohs(CTTUPLE(skb, src.u.all)); +fallback: + return flow_get_proto_src(skb, flow); +} + +static u32 flow_get_nfct_proto_dst(const struct sk_buff *skb, + const struct flow_keys *flow) +{ + return ntohs(CTTUPLE(skb, dst.u.all)); +fallback: + return flow_get_proto_dst(skb, flow); +} + +static u32 flow_get_rtclassid(const struct sk_buff *skb) +{ +#ifdef CONFIG_IP_ROUTE_CLASSID + if (skb_dst(skb)) + return skb_dst(skb)->tclassid; +#endif + return 0; +} + +static u32 flow_get_skuid(const struct sk_buff *skb) +{ + struct sock *sk = skb_to_full_sk(skb); + + if (sk && sk->sk_socket && sk->sk_socket->file) { + kuid_t skuid = sk->sk_socket->file->f_cred->fsuid; + + return from_kuid(&init_user_ns, skuid); + } + return 0; +} + +static u32 flow_get_skgid(const struct sk_buff *skb) +{ + struct sock *sk = skb_to_full_sk(skb); + + if (sk && sk->sk_socket && sk->sk_socket->file) { + kgid_t skgid = sk->sk_socket->file->f_cred->fsgid; + + return from_kgid(&init_user_ns, skgid); + } + return 0; +} + +static u32 flow_get_vlan_tag(const struct sk_buff *skb) +{ + u16 uninitialized_var(tag); + + if (vlan_get_tag(skb, &tag) < 0) + return 0; + return tag & VLAN_VID_MASK; +} + +static u32 flow_get_rxhash(struct sk_buff *skb) +{ + return skb_get_hash(skb); +} + +static u32 flow_key_get(struct sk_buff *skb, int key, struct flow_keys *flow) +{ + switch (key) { + case FLOW_KEY_SRC: + return flow_get_src(skb, flow); + case FLOW_KEY_DST: + return flow_get_dst(skb, flow); + case FLOW_KEY_PROTO: + return flow_get_proto(skb, flow); + case FLOW_KEY_PROTO_SRC: + return flow_get_proto_src(skb, flow); + case FLOW_KEY_PROTO_DST: + return flow_get_proto_dst(skb, flow); + case FLOW_KEY_IIF: + return flow_get_iif(skb); + case FLOW_KEY_PRIORITY: + return flow_get_priority(skb); + case FLOW_KEY_MARK: + return flow_get_mark(skb); + case FLOW_KEY_NFCT: + return flow_get_nfct(skb); + case FLOW_KEY_NFCT_SRC: + return flow_get_nfct_src(skb, flow); + case FLOW_KEY_NFCT_DST: + return flow_get_nfct_dst(skb, flow); + case FLOW_KEY_NFCT_PROTO_SRC: + return flow_get_nfct_proto_src(skb, flow); + case FLOW_KEY_NFCT_PROTO_DST: + return flow_get_nfct_proto_dst(skb, flow); + case FLOW_KEY_RTCLASSID: + return flow_get_rtclassid(skb); + case FLOW_KEY_SKUID: + return flow_get_skuid(skb); + case FLOW_KEY_SKGID: + return flow_get_skgid(skb); + case FLOW_KEY_VLAN_TAG: + return flow_get_vlan_tag(skb); + case FLOW_KEY_RXHASH: + return flow_get_rxhash(skb); + default: + WARN_ON(1); + return 0; + } +} + +#define FLOW_KEYS_NEEDED ((1 << FLOW_KEY_SRC) | \ + (1 << FLOW_KEY_DST) | \ + (1 << FLOW_KEY_PROTO) | \ + (1 << FLOW_KEY_PROTO_SRC) | \ + (1 << FLOW_KEY_PROTO_DST) | \ + (1 << FLOW_KEY_NFCT_SRC) | \ + (1 << FLOW_KEY_NFCT_DST) | \ + (1 << FLOW_KEY_NFCT_PROTO_SRC) | \ + (1 << FLOW_KEY_NFCT_PROTO_DST)) + +static int flow_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct flow_head *head = rcu_dereference_bh(tp->root); + struct flow_filter *f; + u32 keymask; + u32 classid; + unsigned int n, key; + int r; + + list_for_each_entry_rcu(f, &head->filters, list) { + u32 keys[FLOW_KEY_MAX + 1]; + struct flow_keys flow_keys; + + if (!tcf_em_tree_match(skb, &f->ematches, NULL)) + continue; + + keymask = f->keymask; + if (keymask & FLOW_KEYS_NEEDED) + skb_flow_dissect_flow_keys(skb, &flow_keys, 0); + + for (n = 0; n < f->nkeys; n++) { + key = ffs(keymask) - 1; + keymask &= ~(1 << key); + keys[n] = flow_key_get(skb, key, &flow_keys); + } + + if (f->mode == FLOW_MODE_HASH) + classid = jhash2(keys, f->nkeys, f->hashrnd); + else { + classid = keys[0]; + classid = (classid & f->mask) ^ f->xor; + classid = (classid >> f->rshift) + f->addend; + } + + if (f->divisor) + classid %= f->divisor; + + res->class = 0; + res->classid = TC_H_MAKE(f->baseclass, f->baseclass + classid); + + r = tcf_exts_exec(skb, &f->exts, res); + if (r < 0) + continue; + return r; + } + return -1; +} + +static void flow_perturbation(struct timer_list *t) +{ + struct flow_filter *f = from_timer(f, t, perturb_timer); + + get_random_bytes(&f->hashrnd, 4); + if (f->perturb_period) + mod_timer(&f->perturb_timer, jiffies + f->perturb_period); +} + +static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = { + [TCA_FLOW_KEYS] = { .type = NLA_U32 }, + [TCA_FLOW_MODE] = { .type = NLA_U32 }, + [TCA_FLOW_BASECLASS] = { .type = NLA_U32 }, + [TCA_FLOW_RSHIFT] = { .type = NLA_U32 }, + [TCA_FLOW_ADDEND] = { .type = NLA_U32 }, + [TCA_FLOW_MASK] = { .type = NLA_U32 }, + [TCA_FLOW_XOR] = { .type = NLA_U32 }, + [TCA_FLOW_DIVISOR] = { .type = NLA_U32 }, + [TCA_FLOW_ACT] = { .type = NLA_NESTED }, + [TCA_FLOW_POLICE] = { .type = NLA_NESTED }, + [TCA_FLOW_EMATCHES] = { .type = NLA_NESTED }, + [TCA_FLOW_PERTURB] = { .type = NLA_U32 }, +}; + +static void __flow_destroy_filter(struct flow_filter *f) +{ + del_timer_sync(&f->perturb_timer); + tcf_exts_destroy(&f->exts); + tcf_em_tree_destroy(&f->ematches); + tcf_exts_put_net(&f->exts); + kfree(f); +} + +static void flow_destroy_filter_work(struct work_struct *work) +{ + struct flow_filter *f = container_of(to_rcu_work(work), + struct flow_filter, + rwork); + rtnl_lock(); + __flow_destroy_filter(f); + rtnl_unlock(); +} + +static int flow_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + void **arg, bool ovr, struct netlink_ext_ack *extack) +{ + struct flow_head *head = rtnl_dereference(tp->root); + struct flow_filter *fold, *fnew; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_FLOW_MAX + 1]; + unsigned int nkeys = 0; + unsigned int perturb_period = 0; + u32 baseclass = 0; + u32 keymask = 0; + u32 mode; + int err; + + if (opt == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_FLOW_MAX, opt, flow_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_FLOW_BASECLASS]) { + baseclass = nla_get_u32(tb[TCA_FLOW_BASECLASS]); + if (TC_H_MIN(baseclass) == 0) + return -EINVAL; + } + + if (tb[TCA_FLOW_KEYS]) { + keymask = nla_get_u32(tb[TCA_FLOW_KEYS]); + + nkeys = hweight32(keymask); + if (nkeys == 0) + return -EINVAL; + + if (fls(keymask) - 1 > FLOW_KEY_MAX) + return -EOPNOTSUPP; + + if ((keymask & (FLOW_KEY_SKUID|FLOW_KEY_SKGID)) && + sk_user_ns(NETLINK_CB(in_skb).sk) != &init_user_ns) + return -EOPNOTSUPP; + } + + fnew = kzalloc(sizeof(*fnew), GFP_KERNEL); + if (!fnew) + return -ENOBUFS; + + err = tcf_em_tree_validate(tp, tb[TCA_FLOW_EMATCHES], &fnew->ematches); + if (err < 0) + goto err1; + + err = tcf_exts_init(&fnew->exts, TCA_FLOW_ACT, TCA_FLOW_POLICE); + if (err < 0) + goto err2; + + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &fnew->exts, ovr, + extack); + if (err < 0) + goto err2; + + fold = *arg; + if (fold) { + err = -EINVAL; + if (fold->handle != handle && handle) + goto err2; + + /* Copy fold into fnew */ + fnew->tp = fold->tp; + fnew->handle = fold->handle; + fnew->nkeys = fold->nkeys; + fnew->keymask = fold->keymask; + fnew->mode = fold->mode; + fnew->mask = fold->mask; + fnew->xor = fold->xor; + fnew->rshift = fold->rshift; + fnew->addend = fold->addend; + fnew->divisor = fold->divisor; + fnew->baseclass = fold->baseclass; + fnew->hashrnd = fold->hashrnd; + + mode = fold->mode; + if (tb[TCA_FLOW_MODE]) + mode = nla_get_u32(tb[TCA_FLOW_MODE]); + if (mode != FLOW_MODE_HASH && nkeys > 1) + goto err2; + + if (mode == FLOW_MODE_HASH) + perturb_period = fold->perturb_period; + if (tb[TCA_FLOW_PERTURB]) { + if (mode != FLOW_MODE_HASH) + goto err2; + perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ; + } + } else { + err = -EINVAL; + if (!handle) + goto err2; + if (!tb[TCA_FLOW_KEYS]) + goto err2; + + mode = FLOW_MODE_MAP; + if (tb[TCA_FLOW_MODE]) + mode = nla_get_u32(tb[TCA_FLOW_MODE]); + if (mode != FLOW_MODE_HASH && nkeys > 1) + goto err2; + + if (tb[TCA_FLOW_PERTURB]) { + if (mode != FLOW_MODE_HASH) + goto err2; + perturb_period = nla_get_u32(tb[TCA_FLOW_PERTURB]) * HZ; + } + + if (TC_H_MAJ(baseclass) == 0) { + struct Qdisc *q = tcf_block_q(tp->chain->block); + + baseclass = TC_H_MAKE(q->handle, baseclass); + } + if (TC_H_MIN(baseclass) == 0) + baseclass = TC_H_MAKE(baseclass, 1); + + fnew->handle = handle; + fnew->mask = ~0U; + fnew->tp = tp; + get_random_bytes(&fnew->hashrnd, 4); + } + + timer_setup(&fnew->perturb_timer, flow_perturbation, TIMER_DEFERRABLE); + + tcf_block_netif_keep_dst(tp->chain->block); + + if (tb[TCA_FLOW_KEYS]) { + fnew->keymask = keymask; + fnew->nkeys = nkeys; + } + + fnew->mode = mode; + + if (tb[TCA_FLOW_MASK]) + fnew->mask = nla_get_u32(tb[TCA_FLOW_MASK]); + if (tb[TCA_FLOW_XOR]) + fnew->xor = nla_get_u32(tb[TCA_FLOW_XOR]); + if (tb[TCA_FLOW_RSHIFT]) + fnew->rshift = nla_get_u32(tb[TCA_FLOW_RSHIFT]); + if (tb[TCA_FLOW_ADDEND]) + fnew->addend = nla_get_u32(tb[TCA_FLOW_ADDEND]); + + if (tb[TCA_FLOW_DIVISOR]) + fnew->divisor = nla_get_u32(tb[TCA_FLOW_DIVISOR]); + if (baseclass) + fnew->baseclass = baseclass; + + fnew->perturb_period = perturb_period; + if (perturb_period) + mod_timer(&fnew->perturb_timer, jiffies + perturb_period); + + if (!*arg) + list_add_tail_rcu(&fnew->list, &head->filters); + else + list_replace_rcu(&fold->list, &fnew->list); + + *arg = fnew; + + if (fold) { + tcf_exts_get_net(&fold->exts); + tcf_queue_work(&fold->rwork, flow_destroy_filter_work); + } + return 0; + +err2: + tcf_exts_destroy(&fnew->exts); + tcf_em_tree_destroy(&fnew->ematches); +err1: + kfree(fnew); + return err; +} + +static int flow_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct flow_head *head = rtnl_dereference(tp->root); + struct flow_filter *f = arg; + + list_del_rcu(&f->list); + tcf_exts_get_net(&f->exts); + tcf_queue_work(&f->rwork, flow_destroy_filter_work); + *last = list_empty(&head->filters); + return 0; +} + +static int flow_init(struct tcf_proto *tp) +{ + struct flow_head *head; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (head == NULL) + return -ENOBUFS; + INIT_LIST_HEAD(&head->filters); + rcu_assign_pointer(tp->root, head); + return 0; +} + +static void flow_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct flow_head *head = rtnl_dereference(tp->root); + struct flow_filter *f, *next; + + list_for_each_entry_safe(f, next, &head->filters, list) { + list_del_rcu(&f->list); + if (tcf_exts_get_net(&f->exts)) + tcf_queue_work(&f->rwork, flow_destroy_filter_work); + else + __flow_destroy_filter(f); + } + kfree_rcu(head, rcu); +} + +static void *flow_get(struct tcf_proto *tp, u32 handle) +{ + struct flow_head *head = rtnl_dereference(tp->root); + struct flow_filter *f; + + list_for_each_entry(f, &head->filters, list) + if (f->handle == handle) + return f; + return NULL; +} + +static int flow_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct flow_filter *f = fh; + struct nlattr *nest; + + if (f == NULL) + return skb->len; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_FLOW_KEYS, f->keymask) || + nla_put_u32(skb, TCA_FLOW_MODE, f->mode)) + goto nla_put_failure; + + if (f->mask != ~0 || f->xor != 0) { + if (nla_put_u32(skb, TCA_FLOW_MASK, f->mask) || + nla_put_u32(skb, TCA_FLOW_XOR, f->xor)) + goto nla_put_failure; + } + if (f->rshift && + nla_put_u32(skb, TCA_FLOW_RSHIFT, f->rshift)) + goto nla_put_failure; + if (f->addend && + nla_put_u32(skb, TCA_FLOW_ADDEND, f->addend)) + goto nla_put_failure; + + if (f->divisor && + nla_put_u32(skb, TCA_FLOW_DIVISOR, f->divisor)) + goto nla_put_failure; + if (f->baseclass && + nla_put_u32(skb, TCA_FLOW_BASECLASS, f->baseclass)) + goto nla_put_failure; + + if (f->perturb_period && + nla_put_u32(skb, TCA_FLOW_PERTURB, f->perturb_period / HZ)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts) < 0) + goto nla_put_failure; +#ifdef CONFIG_NET_EMATCH + if (f->ematches.hdr.nmatches && + tcf_em_tree_dump(skb, &f->ematches, TCA_FLOW_EMATCHES) < 0) + goto nla_put_failure; +#endif + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void flow_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct flow_head *head = rtnl_dereference(tp->root); + struct flow_filter *f; + + list_for_each_entry(f, &head->filters, list) { + if (arg->count < arg->skip) + goto skip; + if (arg->fn(tp, f, arg) < 0) { + arg->stop = 1; + break; + } +skip: + arg->count++; + } +} + +static struct tcf_proto_ops cls_flow_ops __read_mostly = { + .kind = "flow", + .classify = flow_classify, + .init = flow_init, + .destroy = flow_destroy, + .change = flow_change, + .delete = flow_delete, + .get = flow_get, + .dump = flow_dump, + .walk = flow_walk, + .owner = THIS_MODULE, +}; + +static int __init cls_flow_init(void) +{ + return register_tcf_proto_ops(&cls_flow_ops); +} + +static void __exit cls_flow_exit(void) +{ + unregister_tcf_proto_ops(&cls_flow_ops); +} + +module_init(cls_flow_init); +module_exit(cls_flow_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); +MODULE_DESCRIPTION("TC flow classifier"); diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c new file mode 100644 index 000000000..616364814 --- /dev/null +++ b/net/sched/cls_flower.c @@ -0,0 +1,2001 @@ +/* + * net/sched/cls_flower.c Flower classifier + * + * Copyright (c) 2015 Jiri Pirko <jiri@resnulli.us> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/rhashtable.h> +#include <linux/workqueue.h> + +#include <linux/if_ether.h> +#include <linux/in6.h> +#include <linux/ip.h> +#include <linux/mpls.h> + +#include <net/sch_generic.h> +#include <net/pkt_cls.h> +#include <net/ip.h> +#include <net/flow_dissector.h> +#include <net/geneve.h> + +#include <net/dst.h> +#include <net/dst_metadata.h> + +struct fl_flow_key { + int indev_ifindex; + struct flow_dissector_key_control control; + struct flow_dissector_key_control enc_control; + struct flow_dissector_key_basic basic; + struct flow_dissector_key_eth_addrs eth; + struct flow_dissector_key_vlan vlan; + struct flow_dissector_key_vlan cvlan; + union { + struct flow_dissector_key_ipv4_addrs ipv4; + struct flow_dissector_key_ipv6_addrs ipv6; + }; + struct flow_dissector_key_ports tp; + struct flow_dissector_key_icmp icmp; + struct flow_dissector_key_arp arp; + struct flow_dissector_key_keyid enc_key_id; + union { + struct flow_dissector_key_ipv4_addrs enc_ipv4; + struct flow_dissector_key_ipv6_addrs enc_ipv6; + }; + struct flow_dissector_key_ports enc_tp; + struct flow_dissector_key_mpls mpls; + struct flow_dissector_key_tcp tcp; + struct flow_dissector_key_ip ip; + struct flow_dissector_key_ip enc_ip; + struct flow_dissector_key_enc_opts enc_opts; +} __aligned(BITS_PER_LONG / 8); /* Ensure that we can do comparisons as longs. */ + +struct fl_flow_mask_range { + unsigned short int start; + unsigned short int end; +}; + +struct fl_flow_mask { + struct fl_flow_key key; + struct fl_flow_mask_range range; + struct rhash_head ht_node; + struct rhashtable ht; + struct rhashtable_params filter_ht_params; + struct flow_dissector dissector; + struct list_head filters; + struct rcu_work rwork; + struct list_head list; +}; + +struct fl_flow_tmplt { + struct fl_flow_key dummy_key; + struct fl_flow_key mask; + struct flow_dissector dissector; + struct tcf_chain *chain; +}; + +struct cls_fl_head { + struct rhashtable ht; + struct list_head masks; + struct rcu_work rwork; + struct idr handle_idr; +}; + +struct cls_fl_filter { + struct fl_flow_mask *mask; + struct rhash_head ht_node; + struct fl_flow_key mkey; + struct tcf_exts exts; + struct tcf_result res; + struct fl_flow_key key; + struct list_head list; + u32 handle; + u32 flags; + unsigned int in_hw_count; + struct rcu_work rwork; + struct net_device *hw_dev; +}; + +static const struct rhashtable_params mask_ht_params = { + .key_offset = offsetof(struct fl_flow_mask, key), + .key_len = sizeof(struct fl_flow_key), + .head_offset = offsetof(struct fl_flow_mask, ht_node), + .automatic_shrinking = true, +}; + +static unsigned short int fl_mask_range(const struct fl_flow_mask *mask) +{ + return mask->range.end - mask->range.start; +} + +static void fl_mask_update_range(struct fl_flow_mask *mask) +{ + const u8 *bytes = (const u8 *) &mask->key; + size_t size = sizeof(mask->key); + size_t i, first = 0, last; + + for (i = 0; i < size; i++) { + if (bytes[i]) { + first = i; + break; + } + } + last = first; + for (i = size - 1; i != first; i--) { + if (bytes[i]) { + last = i; + break; + } + } + mask->range.start = rounddown(first, sizeof(long)); + mask->range.end = roundup(last + 1, sizeof(long)); +} + +static void *fl_key_get_start(struct fl_flow_key *key, + const struct fl_flow_mask *mask) +{ + return (u8 *) key + mask->range.start; +} + +static void fl_set_masked_key(struct fl_flow_key *mkey, struct fl_flow_key *key, + struct fl_flow_mask *mask) +{ + const long *lkey = fl_key_get_start(key, mask); + const long *lmask = fl_key_get_start(&mask->key, mask); + long *lmkey = fl_key_get_start(mkey, mask); + int i; + + for (i = 0; i < fl_mask_range(mask); i += sizeof(long)) + *lmkey++ = *lkey++ & *lmask++; +} + +static bool fl_mask_fits_tmplt(struct fl_flow_tmplt *tmplt, + struct fl_flow_mask *mask) +{ + const long *lmask = fl_key_get_start(&mask->key, mask); + const long *ltmplt; + int i; + + if (!tmplt) + return true; + ltmplt = fl_key_get_start(&tmplt->mask, mask); + for (i = 0; i < fl_mask_range(mask); i += sizeof(long)) { + if (~*ltmplt++ & *lmask++) + return false; + } + return true; +} + +static void fl_clear_masked_range(struct fl_flow_key *key, + struct fl_flow_mask *mask) +{ + memset(fl_key_get_start(key, mask), 0, fl_mask_range(mask)); +} + +static struct cls_fl_filter *fl_lookup(struct fl_flow_mask *mask, + struct fl_flow_key *mkey) +{ + return rhashtable_lookup_fast(&mask->ht, fl_key_get_start(mkey, mask), + mask->filter_ht_params); +} + +static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct cls_fl_head *head = rcu_dereference_bh(tp->root); + struct cls_fl_filter *f; + struct fl_flow_mask *mask; + struct fl_flow_key skb_key; + struct fl_flow_key skb_mkey; + + list_for_each_entry_rcu(mask, &head->masks, list) { + flow_dissector_init_keys(&skb_key.control, &skb_key.basic); + fl_clear_masked_range(&skb_key, mask); + + skb_key.indev_ifindex = skb->skb_iif; + /* skb_flow_dissect() does not set n_proto in case an unknown + * protocol, so do it rather here. + */ + skb_key.basic.n_proto = skb_protocol(skb, false); + skb_flow_dissect_tunnel_info(skb, &mask->dissector, &skb_key); + skb_flow_dissect(skb, &mask->dissector, &skb_key, 0); + + fl_set_masked_key(&skb_mkey, &skb_key, mask); + + f = fl_lookup(mask, &skb_mkey); + if (f && !tc_skip_sw(f->flags)) { + *res = f->res; + return tcf_exts_exec(skb, &f->exts, res); + } + } + return -1; +} + +static int fl_init(struct tcf_proto *tp) +{ + struct cls_fl_head *head; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOBUFS; + + INIT_LIST_HEAD_RCU(&head->masks); + rcu_assign_pointer(tp->root, head); + idr_init(&head->handle_idr); + + return rhashtable_init(&head->ht, &mask_ht_params); +} + +static void fl_mask_free(struct fl_flow_mask *mask) +{ + rhashtable_destroy(&mask->ht); + kfree(mask); +} + +static void fl_mask_free_work(struct work_struct *work) +{ + struct fl_flow_mask *mask = container_of(to_rcu_work(work), + struct fl_flow_mask, rwork); + + fl_mask_free(mask); +} + +static bool fl_mask_put(struct cls_fl_head *head, struct fl_flow_mask *mask, + bool async) +{ + if (!list_empty(&mask->filters)) + return false; + + rhashtable_remove_fast(&head->ht, &mask->ht_node, mask_ht_params); + list_del_rcu(&mask->list); + if (async) + tcf_queue_work(&mask->rwork, fl_mask_free_work); + else + fl_mask_free(mask); + + return true; +} + +static void __fl_destroy_filter(struct cls_fl_filter *f) +{ + tcf_exts_destroy(&f->exts); + tcf_exts_put_net(&f->exts); + kfree(f); +} + +static void fl_destroy_filter_work(struct work_struct *work) +{ + struct cls_fl_filter *f = container_of(to_rcu_work(work), + struct cls_fl_filter, rwork); + + rtnl_lock(); + __fl_destroy_filter(f); + rtnl_unlock(); +} + +static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f, + struct netlink_ext_ack *extack) +{ + struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = tp->chain->block; + + tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, extack); + cls_flower.command = TC_CLSFLOWER_DESTROY; + cls_flower.cookie = (unsigned long) f; + + tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, + &cls_flower, false); + tcf_block_offload_dec(block, &f->flags); +} + +static int fl_hw_replace_filter(struct tcf_proto *tp, + struct cls_fl_filter *f, + struct netlink_ext_ack *extack) +{ + struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = tp->chain->block; + bool skip_sw = tc_skip_sw(f->flags); + int err; + + tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, extack); + cls_flower.command = TC_CLSFLOWER_REPLACE; + cls_flower.cookie = (unsigned long) f; + cls_flower.dissector = &f->mask->dissector; + cls_flower.mask = &f->mask->key; + cls_flower.key = &f->mkey; + cls_flower.exts = &f->exts; + cls_flower.classid = f->res.classid; + + err = tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, + &cls_flower, skip_sw); + if (err < 0) { + fl_hw_destroy_filter(tp, f, NULL); + return err; + } else if (err > 0) { + f->in_hw_count = err; + tcf_block_offload_inc(block, &f->flags); + } + + if (skip_sw && !(f->flags & TCA_CLS_FLAGS_IN_HW)) + return -EINVAL; + + return 0; +} + +static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) +{ + struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = tp->chain->block; + + tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, NULL); + cls_flower.command = TC_CLSFLOWER_STATS; + cls_flower.cookie = (unsigned long) f; + cls_flower.exts = &f->exts; + cls_flower.classid = f->res.classid; + + tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, + &cls_flower, false); +} + +static bool __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f, + struct netlink_ext_ack *extack) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + bool async = tcf_exts_get_net(&f->exts); + bool last; + + idr_remove(&head->handle_idr, f->handle); + list_del_rcu(&f->list); + last = fl_mask_put(head, f->mask, async); + if (!tc_skip_hw(f->flags)) + fl_hw_destroy_filter(tp, f, extack); + tcf_unbind_filter(tp, &f->res); + if (async) + tcf_queue_work(&f->rwork, fl_destroy_filter_work); + else + __fl_destroy_filter(f); + + return last; +} + +static void fl_destroy_sleepable(struct work_struct *work) +{ + struct cls_fl_head *head = container_of(to_rcu_work(work), + struct cls_fl_head, + rwork); + + rhashtable_destroy(&head->ht); + kfree(head); + module_put(THIS_MODULE); +} + +static void fl_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + struct fl_flow_mask *mask, *next_mask; + struct cls_fl_filter *f, *next; + + list_for_each_entry_safe(mask, next_mask, &head->masks, list) { + list_for_each_entry_safe(f, next, &mask->filters, list) { + if (__fl_delete(tp, f, extack)) + break; + } + } + idr_destroy(&head->handle_idr); + + __module_get(THIS_MODULE); + tcf_queue_work(&head->rwork, fl_destroy_sleepable); +} + +static void *fl_get(struct tcf_proto *tp, u32 handle) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + + return idr_find(&head->handle_idr, handle); +} + +static const struct nla_policy fl_policy[TCA_FLOWER_MAX + 1] = { + [TCA_FLOWER_UNSPEC] = { .type = NLA_UNSPEC }, + [TCA_FLOWER_CLASSID] = { .type = NLA_U32 }, + [TCA_FLOWER_INDEV] = { .type = NLA_STRING, + .len = IFNAMSIZ }, + [TCA_FLOWER_KEY_ETH_DST] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ETH_DST_MASK] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ETH_SRC] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ETH_SRC_MASK] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ETH_TYPE] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_IP_PROTO] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_IPV4_SRC] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_IPV4_SRC_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_IPV4_DST] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_IPV4_DST_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_IPV6_SRC] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_IPV6_SRC_MASK] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_IPV6_DST] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_IPV6_DST_MASK] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_TCP_SRC] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_TCP_DST] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_UDP_SRC] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_UDP_DST] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_VLAN_ID] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_VLAN_PRIO] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_VLAN_ETH_TYPE] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_KEY_ID] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ENC_IPV4_SRC] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ENC_IPV4_SRC_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ENC_IPV4_DST] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ENC_IPV4_DST_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ENC_IPV6_SRC] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_ENC_IPV6_SRC_MASK] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_ENC_IPV6_DST] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_ENC_IPV6_DST_MASK] = { .len = sizeof(struct in6_addr) }, + [TCA_FLOWER_KEY_TCP_SRC_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_TCP_DST_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_UDP_SRC_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_UDP_DST_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_SCTP_SRC_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_SCTP_DST_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_SCTP_SRC] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_SCTP_DST] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_UDP_SRC_PORT] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_UDP_DST_PORT] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_FLAGS] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_FLAGS_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ICMPV4_TYPE] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV4_TYPE_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV4_CODE] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV4_CODE_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV6_TYPE] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV6_TYPE_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV6_CODE] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ICMPV6_CODE_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ARP_SIP] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ARP_SIP_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ARP_TIP] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ARP_TIP_MASK] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_ARP_OP] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ARP_OP_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ARP_SHA] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ARP_SHA_MASK] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ARP_THA] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_ARP_THA_MASK] = { .len = ETH_ALEN }, + [TCA_FLOWER_KEY_MPLS_TTL] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_MPLS_BOS] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_MPLS_TC] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_MPLS_LABEL] = { .type = NLA_U32 }, + [TCA_FLOWER_KEY_TCP_FLAGS] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_TCP_FLAGS_MASK] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_IP_TOS] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_IP_TOS_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_IP_TTL] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_IP_TTL_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_CVLAN_ID] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_CVLAN_PRIO] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_CVLAN_ETH_TYPE] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_IP_TOS] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ENC_IP_TOS_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ENC_IP_TTL] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ENC_IP_TTL_MASK] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ENC_OPTS] = { .type = NLA_NESTED }, + [TCA_FLOWER_KEY_ENC_OPTS_MASK] = { .type = NLA_NESTED }, + [TCA_FLOWER_FLAGS] = { .type = NLA_U32 }, +}; + +static const struct nla_policy +enc_opts_policy[TCA_FLOWER_KEY_ENC_OPTS_MAX + 1] = { + [TCA_FLOWER_KEY_ENC_OPTS_GENEVE] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy +geneve_opt_policy[TCA_FLOWER_KEY_ENC_OPT_GENEVE_MAX + 1] = { + [TCA_FLOWER_KEY_ENC_OPT_GENEVE_CLASS] = { .type = NLA_U16 }, + [TCA_FLOWER_KEY_ENC_OPT_GENEVE_TYPE] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ENC_OPT_GENEVE_DATA] = { .type = NLA_BINARY, + .len = 128 }, +}; + +static void fl_set_key_val(struct nlattr **tb, + void *val, int val_type, + void *mask, int mask_type, int len) +{ + if (!tb[val_type]) + return; + memcpy(val, nla_data(tb[val_type]), len); + if (mask_type == TCA_FLOWER_UNSPEC || !tb[mask_type]) + memset(mask, 0xff, len); + else + memcpy(mask, nla_data(tb[mask_type]), len); +} + +static int fl_set_key_mpls(struct nlattr **tb, + struct flow_dissector_key_mpls *key_val, + struct flow_dissector_key_mpls *key_mask) +{ + if (tb[TCA_FLOWER_KEY_MPLS_TTL]) { + key_val->mpls_ttl = nla_get_u8(tb[TCA_FLOWER_KEY_MPLS_TTL]); + key_mask->mpls_ttl = MPLS_TTL_MASK; + } + if (tb[TCA_FLOWER_KEY_MPLS_BOS]) { + u8 bos = nla_get_u8(tb[TCA_FLOWER_KEY_MPLS_BOS]); + + if (bos & ~MPLS_BOS_MASK) + return -EINVAL; + key_val->mpls_bos = bos; + key_mask->mpls_bos = MPLS_BOS_MASK; + } + if (tb[TCA_FLOWER_KEY_MPLS_TC]) { + u8 tc = nla_get_u8(tb[TCA_FLOWER_KEY_MPLS_TC]); + + if (tc & ~MPLS_TC_MASK) + return -EINVAL; + key_val->mpls_tc = tc; + key_mask->mpls_tc = MPLS_TC_MASK; + } + if (tb[TCA_FLOWER_KEY_MPLS_LABEL]) { + u32 label = nla_get_u32(tb[TCA_FLOWER_KEY_MPLS_LABEL]); + + if (label & ~MPLS_LABEL_MASK) + return -EINVAL; + key_val->mpls_label = label; + key_mask->mpls_label = MPLS_LABEL_MASK; + } + return 0; +} + +static void fl_set_key_vlan(struct nlattr **tb, + __be16 ethertype, + int vlan_id_key, int vlan_prio_key, + int vlan_next_eth_type_key, + struct flow_dissector_key_vlan *key_val, + struct flow_dissector_key_vlan *key_mask) +{ +#define VLAN_PRIORITY_MASK 0x7 + + if (tb[vlan_id_key]) { + key_val->vlan_id = + nla_get_u16(tb[vlan_id_key]) & VLAN_VID_MASK; + key_mask->vlan_id = VLAN_VID_MASK; + } + if (tb[vlan_prio_key]) { + key_val->vlan_priority = + nla_get_u8(tb[vlan_prio_key]) & + VLAN_PRIORITY_MASK; + key_mask->vlan_priority = VLAN_PRIORITY_MASK; + } + key_val->vlan_tpid = ethertype; + key_mask->vlan_tpid = cpu_to_be16(~0); + if (tb[vlan_next_eth_type_key]) { + key_val->vlan_eth_type = + nla_get_be16(tb[vlan_next_eth_type_key]); + key_mask->vlan_eth_type = cpu_to_be16(~0); + } +} + +static void fl_set_key_flag(u32 flower_key, u32 flower_mask, + u32 *dissector_key, u32 *dissector_mask, + u32 flower_flag_bit, u32 dissector_flag_bit) +{ + if (flower_mask & flower_flag_bit) { + *dissector_mask |= dissector_flag_bit; + if (flower_key & flower_flag_bit) + *dissector_key |= dissector_flag_bit; + } +} + +static int fl_set_key_flags(struct nlattr **tb, + u32 *flags_key, u32 *flags_mask) +{ + u32 key, mask; + + /* mask is mandatory for flags */ + if (!tb[TCA_FLOWER_KEY_FLAGS_MASK]) + return -EINVAL; + + key = be32_to_cpu(nla_get_u32(tb[TCA_FLOWER_KEY_FLAGS])); + mask = be32_to_cpu(nla_get_u32(tb[TCA_FLOWER_KEY_FLAGS_MASK])); + + *flags_key = 0; + *flags_mask = 0; + + fl_set_key_flag(key, mask, flags_key, flags_mask, + TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT, FLOW_DIS_IS_FRAGMENT); + fl_set_key_flag(key, mask, flags_key, flags_mask, + TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST, + FLOW_DIS_FIRST_FRAG); + + return 0; +} + +static void fl_set_key_ip(struct nlattr **tb, bool encap, + struct flow_dissector_key_ip *key, + struct flow_dissector_key_ip *mask) +{ + int tos_key = encap ? TCA_FLOWER_KEY_ENC_IP_TOS : TCA_FLOWER_KEY_IP_TOS; + int ttl_key = encap ? TCA_FLOWER_KEY_ENC_IP_TTL : TCA_FLOWER_KEY_IP_TTL; + int tos_mask = encap ? TCA_FLOWER_KEY_ENC_IP_TOS_MASK : TCA_FLOWER_KEY_IP_TOS_MASK; + int ttl_mask = encap ? TCA_FLOWER_KEY_ENC_IP_TTL_MASK : TCA_FLOWER_KEY_IP_TTL_MASK; + + fl_set_key_val(tb, &key->tos, tos_key, &mask->tos, tos_mask, sizeof(key->tos)); + fl_set_key_val(tb, &key->ttl, ttl_key, &mask->ttl, ttl_mask, sizeof(key->ttl)); +} + +static int fl_set_geneve_opt(const struct nlattr *nla, struct fl_flow_key *key, + int depth, int option_len, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_MAX + 1]; + struct nlattr *class = NULL, *type = NULL, *data = NULL; + struct geneve_opt *opt; + int err, data_len = 0; + + if (option_len > sizeof(struct geneve_opt)) + data_len = option_len - sizeof(struct geneve_opt); + + opt = (struct geneve_opt *)&key->enc_opts.data[key->enc_opts.len]; + memset(opt, 0xff, option_len); + opt->length = data_len / 4; + opt->r1 = 0; + opt->r2 = 0; + opt->r3 = 0; + + /* If no mask has been prodived we assume an exact match. */ + if (!depth) + return sizeof(struct geneve_opt) + data_len; + + if (nla_type(nla) != TCA_FLOWER_KEY_ENC_OPTS_GENEVE) { + NL_SET_ERR_MSG(extack, "Non-geneve option type for mask"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_FLOWER_KEY_ENC_OPT_GENEVE_MAX, + nla, geneve_opt_policy, extack); + if (err < 0) + return err; + + /* We are not allowed to omit any of CLASS, TYPE or DATA + * fields from the key. + */ + if (!option_len && + (!tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_CLASS] || + !tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_TYPE] || + !tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_DATA])) { + NL_SET_ERR_MSG(extack, "Missing tunnel key geneve option class, type or data"); + return -EINVAL; + } + + /* Omitting any of CLASS, TYPE or DATA fields is allowed + * for the mask. + */ + if (tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_DATA]) { + int new_len = key->enc_opts.len; + + data = tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_DATA]; + data_len = nla_len(data); + if (data_len < 4) { + NL_SET_ERR_MSG(extack, "Tunnel key geneve option data is less than 4 bytes long"); + return -ERANGE; + } + if (data_len % 4) { + NL_SET_ERR_MSG(extack, "Tunnel key geneve option data is not a multiple of 4 bytes long"); + return -ERANGE; + } + + new_len += sizeof(struct geneve_opt) + data_len; + BUILD_BUG_ON(FLOW_DIS_TUN_OPTS_MAX != IP_TUNNEL_OPTS_MAX); + if (new_len > FLOW_DIS_TUN_OPTS_MAX) { + NL_SET_ERR_MSG(extack, "Tunnel options exceeds max size"); + return -ERANGE; + } + opt->length = data_len / 4; + memcpy(opt->opt_data, nla_data(data), data_len); + } + + if (tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_CLASS]) { + class = tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_CLASS]; + opt->opt_class = nla_get_be16(class); + } + + if (tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_TYPE]) { + type = tb[TCA_FLOWER_KEY_ENC_OPT_GENEVE_TYPE]; + opt->type = nla_get_u8(type); + } + + return sizeof(struct geneve_opt) + data_len; +} + +static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, + struct fl_flow_key *mask, + struct netlink_ext_ack *extack) +{ + const struct nlattr *nla_enc_key, *nla_opt_key, *nla_opt_msk = NULL; + int err, option_len, key_depth, msk_depth = 0; + + err = nla_validate_nested(tb[TCA_FLOWER_KEY_ENC_OPTS], + TCA_FLOWER_KEY_ENC_OPTS_MAX, + enc_opts_policy, extack); + if (err) + return err; + + nla_enc_key = nla_data(tb[TCA_FLOWER_KEY_ENC_OPTS]); + + if (tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]) { + err = nla_validate_nested(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK], + TCA_FLOWER_KEY_ENC_OPTS_MAX, + enc_opts_policy, extack); + if (err) + return err; + + nla_opt_msk = nla_data(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]); + msk_depth = nla_len(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]); + } + + nla_for_each_attr(nla_opt_key, nla_enc_key, + nla_len(tb[TCA_FLOWER_KEY_ENC_OPTS]), key_depth) { + switch (nla_type(nla_opt_key)) { + case TCA_FLOWER_KEY_ENC_OPTS_GENEVE: + option_len = 0; + key->enc_opts.dst_opt_type = TUNNEL_GENEVE_OPT; + option_len = fl_set_geneve_opt(nla_opt_key, key, + key_depth, option_len, + extack); + if (option_len < 0) + return option_len; + + key->enc_opts.len += option_len; + /* At the same time we need to parse through the mask + * in order to verify exact and mask attribute lengths. + */ + mask->enc_opts.dst_opt_type = TUNNEL_GENEVE_OPT; + option_len = fl_set_geneve_opt(nla_opt_msk, mask, + msk_depth, option_len, + extack); + if (option_len < 0) + return option_len; + + mask->enc_opts.len += option_len; + if (key->enc_opts.len != mask->enc_opts.len) { + NL_SET_ERR_MSG(extack, "Key and mask miss aligned"); + return -EINVAL; + } + + if (msk_depth) + nla_opt_msk = nla_next(nla_opt_msk, &msk_depth); + break; + default: + NL_SET_ERR_MSG(extack, "Unknown tunnel option type"); + return -EINVAL; + } + } + + return 0; +} + +static int fl_set_key(struct net *net, struct nlattr **tb, + struct fl_flow_key *key, struct fl_flow_key *mask, + struct netlink_ext_ack *extack) +{ + __be16 ethertype; + int ret = 0; +#ifdef CONFIG_NET_CLS_IND + if (tb[TCA_FLOWER_INDEV]) { + int err = tcf_change_indev(net, tb[TCA_FLOWER_INDEV], extack); + if (err < 0) + return err; + key->indev_ifindex = err; + mask->indev_ifindex = 0xffffffff; + } +#endif + + fl_set_key_val(tb, key->eth.dst, TCA_FLOWER_KEY_ETH_DST, + mask->eth.dst, TCA_FLOWER_KEY_ETH_DST_MASK, + sizeof(key->eth.dst)); + fl_set_key_val(tb, key->eth.src, TCA_FLOWER_KEY_ETH_SRC, + mask->eth.src, TCA_FLOWER_KEY_ETH_SRC_MASK, + sizeof(key->eth.src)); + + if (tb[TCA_FLOWER_KEY_ETH_TYPE]) { + ethertype = nla_get_be16(tb[TCA_FLOWER_KEY_ETH_TYPE]); + + if (eth_type_vlan(ethertype)) { + fl_set_key_vlan(tb, ethertype, TCA_FLOWER_KEY_VLAN_ID, + TCA_FLOWER_KEY_VLAN_PRIO, + TCA_FLOWER_KEY_VLAN_ETH_TYPE, + &key->vlan, &mask->vlan); + + if (tb[TCA_FLOWER_KEY_VLAN_ETH_TYPE]) { + ethertype = nla_get_be16(tb[TCA_FLOWER_KEY_VLAN_ETH_TYPE]); + if (eth_type_vlan(ethertype)) { + fl_set_key_vlan(tb, ethertype, + TCA_FLOWER_KEY_CVLAN_ID, + TCA_FLOWER_KEY_CVLAN_PRIO, + TCA_FLOWER_KEY_CVLAN_ETH_TYPE, + &key->cvlan, &mask->cvlan); + fl_set_key_val(tb, &key->basic.n_proto, + TCA_FLOWER_KEY_CVLAN_ETH_TYPE, + &mask->basic.n_proto, + TCA_FLOWER_UNSPEC, + sizeof(key->basic.n_proto)); + } else { + key->basic.n_proto = ethertype; + mask->basic.n_proto = cpu_to_be16(~0); + } + } + } else { + key->basic.n_proto = ethertype; + mask->basic.n_proto = cpu_to_be16(~0); + } + } + + if (key->basic.n_proto == htons(ETH_P_IP) || + key->basic.n_proto == htons(ETH_P_IPV6)) { + fl_set_key_val(tb, &key->basic.ip_proto, TCA_FLOWER_KEY_IP_PROTO, + &mask->basic.ip_proto, TCA_FLOWER_UNSPEC, + sizeof(key->basic.ip_proto)); + fl_set_key_ip(tb, false, &key->ip, &mask->ip); + } + + if (tb[TCA_FLOWER_KEY_IPV4_SRC] || tb[TCA_FLOWER_KEY_IPV4_DST]) { + key->control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; + mask->control.addr_type = ~0; + fl_set_key_val(tb, &key->ipv4.src, TCA_FLOWER_KEY_IPV4_SRC, + &mask->ipv4.src, TCA_FLOWER_KEY_IPV4_SRC_MASK, + sizeof(key->ipv4.src)); + fl_set_key_val(tb, &key->ipv4.dst, TCA_FLOWER_KEY_IPV4_DST, + &mask->ipv4.dst, TCA_FLOWER_KEY_IPV4_DST_MASK, + sizeof(key->ipv4.dst)); + } else if (tb[TCA_FLOWER_KEY_IPV6_SRC] || tb[TCA_FLOWER_KEY_IPV6_DST]) { + key->control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; + mask->control.addr_type = ~0; + fl_set_key_val(tb, &key->ipv6.src, TCA_FLOWER_KEY_IPV6_SRC, + &mask->ipv6.src, TCA_FLOWER_KEY_IPV6_SRC_MASK, + sizeof(key->ipv6.src)); + fl_set_key_val(tb, &key->ipv6.dst, TCA_FLOWER_KEY_IPV6_DST, + &mask->ipv6.dst, TCA_FLOWER_KEY_IPV6_DST_MASK, + sizeof(key->ipv6.dst)); + } + + if (key->basic.ip_proto == IPPROTO_TCP) { + fl_set_key_val(tb, &key->tp.src, TCA_FLOWER_KEY_TCP_SRC, + &mask->tp.src, TCA_FLOWER_KEY_TCP_SRC_MASK, + sizeof(key->tp.src)); + fl_set_key_val(tb, &key->tp.dst, TCA_FLOWER_KEY_TCP_DST, + &mask->tp.dst, TCA_FLOWER_KEY_TCP_DST_MASK, + sizeof(key->tp.dst)); + fl_set_key_val(tb, &key->tcp.flags, TCA_FLOWER_KEY_TCP_FLAGS, + &mask->tcp.flags, TCA_FLOWER_KEY_TCP_FLAGS_MASK, + sizeof(key->tcp.flags)); + } else if (key->basic.ip_proto == IPPROTO_UDP) { + fl_set_key_val(tb, &key->tp.src, TCA_FLOWER_KEY_UDP_SRC, + &mask->tp.src, TCA_FLOWER_KEY_UDP_SRC_MASK, + sizeof(key->tp.src)); + fl_set_key_val(tb, &key->tp.dst, TCA_FLOWER_KEY_UDP_DST, + &mask->tp.dst, TCA_FLOWER_KEY_UDP_DST_MASK, + sizeof(key->tp.dst)); + } else if (key->basic.ip_proto == IPPROTO_SCTP) { + fl_set_key_val(tb, &key->tp.src, TCA_FLOWER_KEY_SCTP_SRC, + &mask->tp.src, TCA_FLOWER_KEY_SCTP_SRC_MASK, + sizeof(key->tp.src)); + fl_set_key_val(tb, &key->tp.dst, TCA_FLOWER_KEY_SCTP_DST, + &mask->tp.dst, TCA_FLOWER_KEY_SCTP_DST_MASK, + sizeof(key->tp.dst)); + } else if (key->basic.n_proto == htons(ETH_P_IP) && + key->basic.ip_proto == IPPROTO_ICMP) { + fl_set_key_val(tb, &key->icmp.type, TCA_FLOWER_KEY_ICMPV4_TYPE, + &mask->icmp.type, + TCA_FLOWER_KEY_ICMPV4_TYPE_MASK, + sizeof(key->icmp.type)); + fl_set_key_val(tb, &key->icmp.code, TCA_FLOWER_KEY_ICMPV4_CODE, + &mask->icmp.code, + TCA_FLOWER_KEY_ICMPV4_CODE_MASK, + sizeof(key->icmp.code)); + } else if (key->basic.n_proto == htons(ETH_P_IPV6) && + key->basic.ip_proto == IPPROTO_ICMPV6) { + fl_set_key_val(tb, &key->icmp.type, TCA_FLOWER_KEY_ICMPV6_TYPE, + &mask->icmp.type, + TCA_FLOWER_KEY_ICMPV6_TYPE_MASK, + sizeof(key->icmp.type)); + fl_set_key_val(tb, &key->icmp.code, TCA_FLOWER_KEY_ICMPV6_CODE, + &mask->icmp.code, + TCA_FLOWER_KEY_ICMPV6_CODE_MASK, + sizeof(key->icmp.code)); + } else if (key->basic.n_proto == htons(ETH_P_MPLS_UC) || + key->basic.n_proto == htons(ETH_P_MPLS_MC)) { + ret = fl_set_key_mpls(tb, &key->mpls, &mask->mpls); + if (ret) + return ret; + } else if (key->basic.n_proto == htons(ETH_P_ARP) || + key->basic.n_proto == htons(ETH_P_RARP)) { + fl_set_key_val(tb, &key->arp.sip, TCA_FLOWER_KEY_ARP_SIP, + &mask->arp.sip, TCA_FLOWER_KEY_ARP_SIP_MASK, + sizeof(key->arp.sip)); + fl_set_key_val(tb, &key->arp.tip, TCA_FLOWER_KEY_ARP_TIP, + &mask->arp.tip, TCA_FLOWER_KEY_ARP_TIP_MASK, + sizeof(key->arp.tip)); + fl_set_key_val(tb, &key->arp.op, TCA_FLOWER_KEY_ARP_OP, + &mask->arp.op, TCA_FLOWER_KEY_ARP_OP_MASK, + sizeof(key->arp.op)); + fl_set_key_val(tb, key->arp.sha, TCA_FLOWER_KEY_ARP_SHA, + mask->arp.sha, TCA_FLOWER_KEY_ARP_SHA_MASK, + sizeof(key->arp.sha)); + fl_set_key_val(tb, key->arp.tha, TCA_FLOWER_KEY_ARP_THA, + mask->arp.tha, TCA_FLOWER_KEY_ARP_THA_MASK, + sizeof(key->arp.tha)); + } + + if (tb[TCA_FLOWER_KEY_ENC_IPV4_SRC] || + tb[TCA_FLOWER_KEY_ENC_IPV4_DST]) { + key->enc_control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS; + mask->enc_control.addr_type = ~0; + fl_set_key_val(tb, &key->enc_ipv4.src, + TCA_FLOWER_KEY_ENC_IPV4_SRC, + &mask->enc_ipv4.src, + TCA_FLOWER_KEY_ENC_IPV4_SRC_MASK, + sizeof(key->enc_ipv4.src)); + fl_set_key_val(tb, &key->enc_ipv4.dst, + TCA_FLOWER_KEY_ENC_IPV4_DST, + &mask->enc_ipv4.dst, + TCA_FLOWER_KEY_ENC_IPV4_DST_MASK, + sizeof(key->enc_ipv4.dst)); + } + + if (tb[TCA_FLOWER_KEY_ENC_IPV6_SRC] || + tb[TCA_FLOWER_KEY_ENC_IPV6_DST]) { + key->enc_control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS; + mask->enc_control.addr_type = ~0; + fl_set_key_val(tb, &key->enc_ipv6.src, + TCA_FLOWER_KEY_ENC_IPV6_SRC, + &mask->enc_ipv6.src, + TCA_FLOWER_KEY_ENC_IPV6_SRC_MASK, + sizeof(key->enc_ipv6.src)); + fl_set_key_val(tb, &key->enc_ipv6.dst, + TCA_FLOWER_KEY_ENC_IPV6_DST, + &mask->enc_ipv6.dst, + TCA_FLOWER_KEY_ENC_IPV6_DST_MASK, + sizeof(key->enc_ipv6.dst)); + } + + fl_set_key_val(tb, &key->enc_key_id.keyid, TCA_FLOWER_KEY_ENC_KEY_ID, + &mask->enc_key_id.keyid, TCA_FLOWER_UNSPEC, + sizeof(key->enc_key_id.keyid)); + + fl_set_key_val(tb, &key->enc_tp.src, TCA_FLOWER_KEY_ENC_UDP_SRC_PORT, + &mask->enc_tp.src, TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK, + sizeof(key->enc_tp.src)); + + fl_set_key_val(tb, &key->enc_tp.dst, TCA_FLOWER_KEY_ENC_UDP_DST_PORT, + &mask->enc_tp.dst, TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK, + sizeof(key->enc_tp.dst)); + + fl_set_key_ip(tb, true, &key->enc_ip, &mask->enc_ip); + + if (tb[TCA_FLOWER_KEY_ENC_OPTS]) { + ret = fl_set_enc_opt(tb, key, mask, extack); + if (ret) + return ret; + } + + if (tb[TCA_FLOWER_KEY_FLAGS]) + ret = fl_set_key_flags(tb, &key->control.flags, &mask->control.flags); + + return ret; +} + +static void fl_mask_copy(struct fl_flow_mask *dst, + struct fl_flow_mask *src) +{ + const void *psrc = fl_key_get_start(&src->key, src); + void *pdst = fl_key_get_start(&dst->key, src); + + memcpy(pdst, psrc, fl_mask_range(src)); + dst->range = src->range; +} + +static const struct rhashtable_params fl_ht_params = { + .key_offset = offsetof(struct cls_fl_filter, mkey), /* base offset */ + .head_offset = offsetof(struct cls_fl_filter, ht_node), + .automatic_shrinking = true, +}; + +static int fl_init_mask_hashtable(struct fl_flow_mask *mask) +{ + mask->filter_ht_params = fl_ht_params; + mask->filter_ht_params.key_len = fl_mask_range(mask); + mask->filter_ht_params.key_offset += mask->range.start; + + return rhashtable_init(&mask->ht, &mask->filter_ht_params); +} + +#define FL_KEY_MEMBER_OFFSET(member) offsetof(struct fl_flow_key, member) +#define FL_KEY_MEMBER_SIZE(member) (sizeof(((struct fl_flow_key *) 0)->member)) + +#define FL_KEY_IS_MASKED(mask, member) \ + memchr_inv(((char *)mask) + FL_KEY_MEMBER_OFFSET(member), \ + 0, FL_KEY_MEMBER_SIZE(member)) \ + +#define FL_KEY_SET(keys, cnt, id, member) \ + do { \ + keys[cnt].key_id = id; \ + keys[cnt].offset = FL_KEY_MEMBER_OFFSET(member); \ + cnt++; \ + } while(0); + +#define FL_KEY_SET_IF_MASKED(mask, keys, cnt, id, member) \ + do { \ + if (FL_KEY_IS_MASKED(mask, member)) \ + FL_KEY_SET(keys, cnt, id, member); \ + } while(0); + +static void fl_init_dissector(struct flow_dissector *dissector, + struct fl_flow_key *mask) +{ + struct flow_dissector_key keys[FLOW_DISSECTOR_KEY_MAX]; + size_t cnt = 0; + + FL_KEY_SET(keys, cnt, FLOW_DISSECTOR_KEY_CONTROL, control); + FL_KEY_SET(keys, cnt, FLOW_DISSECTOR_KEY_BASIC, basic); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ETH_ADDRS, eth); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_IPV4_ADDRS, ipv4); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_IPV6_ADDRS, ipv6); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_PORTS, tp); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_IP, ip); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_TCP, tcp); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ICMP, icmp); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ARP, arp); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_MPLS, mpls); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_VLAN, vlan); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_CVLAN, cvlan); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ENC_KEYID, enc_key_id); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS, enc_ipv4); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS, enc_ipv6); + if (FL_KEY_IS_MASKED(mask, enc_ipv4) || + FL_KEY_IS_MASKED(mask, enc_ipv6)) + FL_KEY_SET(keys, cnt, FLOW_DISSECTOR_KEY_ENC_CONTROL, + enc_control); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ENC_PORTS, enc_tp); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ENC_IP, enc_ip); + FL_KEY_SET_IF_MASKED(mask, keys, cnt, + FLOW_DISSECTOR_KEY_ENC_OPTS, enc_opts); + + skb_flow_dissector_init(dissector, keys, cnt); +} + +static struct fl_flow_mask *fl_create_new_mask(struct cls_fl_head *head, + struct fl_flow_mask *mask) +{ + struct fl_flow_mask *newmask; + int err; + + newmask = kzalloc(sizeof(*newmask), GFP_KERNEL); + if (!newmask) + return ERR_PTR(-ENOMEM); + + fl_mask_copy(newmask, mask); + + err = fl_init_mask_hashtable(newmask); + if (err) + goto errout_free; + + fl_init_dissector(&newmask->dissector, &newmask->key); + + INIT_LIST_HEAD_RCU(&newmask->filters); + + err = rhashtable_insert_fast(&head->ht, &newmask->ht_node, + mask_ht_params); + if (err) + goto errout_destroy; + + list_add_tail_rcu(&newmask->list, &head->masks); + + return newmask; + +errout_destroy: + rhashtable_destroy(&newmask->ht); +errout_free: + kfree(newmask); + + return ERR_PTR(err); +} + +static int fl_check_assign_mask(struct cls_fl_head *head, + struct cls_fl_filter *fnew, + struct cls_fl_filter *fold, + struct fl_flow_mask *mask) +{ + struct fl_flow_mask *newmask; + + fnew->mask = rhashtable_lookup_fast(&head->ht, mask, mask_ht_params); + if (!fnew->mask) { + if (fold) + return -EINVAL; + + newmask = fl_create_new_mask(head, mask); + if (IS_ERR(newmask)) + return PTR_ERR(newmask); + + fnew->mask = newmask; + } else if (fold && fold->mask != fnew->mask) { + return -EINVAL; + } + + return 0; +} + +static int fl_set_parms(struct net *net, struct tcf_proto *tp, + struct cls_fl_filter *f, struct fl_flow_mask *mask, + unsigned long base, struct nlattr **tb, + struct nlattr *est, bool ovr, + struct fl_flow_tmplt *tmplt, + struct netlink_ext_ack *extack) +{ + int err; + + err = tcf_exts_validate(net, tp, tb, est, &f->exts, ovr, extack); + if (err < 0) + return err; + + if (tb[TCA_FLOWER_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_FLOWER_CLASSID]); + tcf_bind_filter(tp, &f->res, base); + } + + err = fl_set_key(net, tb, &f->key, &mask->key, extack); + if (err) + return err; + + fl_mask_update_range(mask); + fl_set_masked_key(&f->mkey, &f->key, mask); + + if (!fl_mask_fits_tmplt(tmplt, mask)) { + NL_SET_ERR_MSG_MOD(extack, "Mask does not fit the template"); + return -EINVAL; + } + + return 0; +} + +static int fl_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + void **arg, bool ovr, struct netlink_ext_ack *extack) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + struct cls_fl_filter *fold = *arg; + struct cls_fl_filter *fnew; + struct fl_flow_mask *mask; + struct nlattr **tb; + int err; + + if (!tca[TCA_OPTIONS]) + return -EINVAL; + + mask = kzalloc(sizeof(struct fl_flow_mask), GFP_KERNEL); + if (!mask) + return -ENOBUFS; + + tb = kcalloc(TCA_FLOWER_MAX + 1, sizeof(struct nlattr *), GFP_KERNEL); + if (!tb) { + err = -ENOBUFS; + goto errout_mask_alloc; + } + + err = nla_parse_nested(tb, TCA_FLOWER_MAX, tca[TCA_OPTIONS], + fl_policy, NULL); + if (err < 0) + goto errout_tb; + + if (fold && handle && fold->handle != handle) { + err = -EINVAL; + goto errout_tb; + } + + fnew = kzalloc(sizeof(*fnew), GFP_KERNEL); + if (!fnew) { + err = -ENOBUFS; + goto errout_tb; + } + + err = tcf_exts_init(&fnew->exts, TCA_FLOWER_ACT, 0); + if (err < 0) + goto errout; + + if (tb[TCA_FLOWER_FLAGS]) { + fnew->flags = nla_get_u32(tb[TCA_FLOWER_FLAGS]); + + if (!tc_flags_valid(fnew->flags)) { + err = -EINVAL; + goto errout; + } + } + + err = fl_set_parms(net, tp, fnew, mask, base, tb, tca[TCA_RATE], ovr, + tp->chain->tmplt_priv, extack); + if (err) + goto errout; + + err = fl_check_assign_mask(head, fnew, fold, mask); + if (err) + goto errout; + + if (!handle) { + handle = 1; + err = idr_alloc_u32(&head->handle_idr, fnew, &handle, + INT_MAX, GFP_KERNEL); + } else if (!fold) { + /* user specifies a handle and it doesn't exist */ + err = idr_alloc_u32(&head->handle_idr, fnew, &handle, + handle, GFP_KERNEL); + } + if (err) + goto errout_mask; + fnew->handle = handle; + + if (!tc_skip_sw(fnew->flags)) { + if (!fold && fl_lookup(fnew->mask, &fnew->mkey)) { + err = -EEXIST; + goto errout_idr; + } + + err = rhashtable_insert_fast(&fnew->mask->ht, &fnew->ht_node, + fnew->mask->filter_ht_params); + if (err) + goto errout_idr; + } + + if (!tc_skip_hw(fnew->flags)) { + err = fl_hw_replace_filter(tp, fnew, extack); + if (err) + goto errout_mask; + } + + if (!tc_in_hw(fnew->flags)) + fnew->flags |= TCA_CLS_FLAGS_NOT_IN_HW; + + if (fold) { + if (!tc_skip_sw(fold->flags)) + rhashtable_remove_fast(&fold->mask->ht, + &fold->ht_node, + fold->mask->filter_ht_params); + if (!tc_skip_hw(fold->flags)) + fl_hw_destroy_filter(tp, fold, NULL); + } + + *arg = fnew; + + if (fold) { + idr_replace(&head->handle_idr, fnew, fnew->handle); + list_replace_rcu(&fold->list, &fnew->list); + tcf_unbind_filter(tp, &fold->res); + tcf_exts_get_net(&fold->exts); + tcf_queue_work(&fold->rwork, fl_destroy_filter_work); + } else { + list_add_tail_rcu(&fnew->list, &fnew->mask->filters); + } + + kfree(tb); + kfree(mask); + return 0; + +errout_idr: + if (!fold) + idr_remove(&head->handle_idr, fnew->handle); + +errout_mask: + fl_mask_put(head, fnew->mask, false); + +errout: + tcf_exts_destroy(&fnew->exts); + kfree(fnew); +errout_tb: + kfree(tb); +errout_mask_alloc: + kfree(mask); + return err; +} + +static int fl_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + struct cls_fl_filter *f = arg; + + if (!tc_skip_sw(f->flags)) + rhashtable_remove_fast(&f->mask->ht, &f->ht_node, + f->mask->filter_ht_params); + __fl_delete(tp, f, extack); + *last = list_empty(&head->masks); + return 0; +} + +static void fl_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + struct cls_fl_filter *f; + + arg->count = arg->skip; + + while ((f = idr_get_next_ul(&head->handle_idr, + &arg->cookie)) != NULL) { + if (arg->fn(tp, f, arg) < 0) { + arg->stop = 1; + break; + } + arg->cookie = f->handle + 1; + arg->count++; + } +} + +static int fl_reoffload(struct tcf_proto *tp, bool add, tc_setup_cb_t *cb, + void *cb_priv, struct netlink_ext_ack *extack) +{ + struct cls_fl_head *head = rtnl_dereference(tp->root); + struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = tp->chain->block; + struct fl_flow_mask *mask; + struct cls_fl_filter *f; + int err; + + list_for_each_entry(mask, &head->masks, list) { + list_for_each_entry(f, &mask->filters, list) { + if (tc_skip_hw(f->flags)) + continue; + + tc_cls_common_offload_init(&cls_flower.common, tp, + f->flags, extack); + cls_flower.command = add ? + TC_CLSFLOWER_REPLACE : TC_CLSFLOWER_DESTROY; + cls_flower.cookie = (unsigned long)f; + cls_flower.dissector = &mask->dissector; + cls_flower.mask = &mask->key; + cls_flower.key = &f->mkey; + cls_flower.exts = &f->exts; + cls_flower.classid = f->res.classid; + + err = cb(TC_SETUP_CLSFLOWER, &cls_flower, cb_priv); + if (err) { + if (add && tc_skip_sw(f->flags)) + return err; + continue; + } + + tc_cls_offload_cnt_update(block, &f->in_hw_count, + &f->flags, add); + } + } + + return 0; +} + +static void fl_hw_create_tmplt(struct tcf_chain *chain, + struct fl_flow_tmplt *tmplt) +{ + struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = chain->block; + struct tcf_exts dummy_exts = { 0, }; + + cls_flower.common.chain_index = chain->index; + cls_flower.command = TC_CLSFLOWER_TMPLT_CREATE; + cls_flower.cookie = (unsigned long) tmplt; + cls_flower.dissector = &tmplt->dissector; + cls_flower.mask = &tmplt->mask; + cls_flower.key = &tmplt->dummy_key; + cls_flower.exts = &dummy_exts; + + /* We don't care if driver (any of them) fails to handle this + * call. It serves just as a hint for it. + */ + tc_setup_cb_call(block, NULL, TC_SETUP_CLSFLOWER, + &cls_flower, false); +} + +static void fl_hw_destroy_tmplt(struct tcf_chain *chain, + struct fl_flow_tmplt *tmplt) +{ + struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = chain->block; + + cls_flower.common.chain_index = chain->index; + cls_flower.command = TC_CLSFLOWER_TMPLT_DESTROY; + cls_flower.cookie = (unsigned long) tmplt; + + tc_setup_cb_call(block, NULL, TC_SETUP_CLSFLOWER, + &cls_flower, false); +} + +static void *fl_tmplt_create(struct net *net, struct tcf_chain *chain, + struct nlattr **tca, + struct netlink_ext_ack *extack) +{ + struct fl_flow_tmplt *tmplt; + struct nlattr **tb; + int err; + + if (!tca[TCA_OPTIONS]) + return ERR_PTR(-EINVAL); + + tb = kcalloc(TCA_FLOWER_MAX + 1, sizeof(struct nlattr *), GFP_KERNEL); + if (!tb) + return ERR_PTR(-ENOBUFS); + err = nla_parse_nested(tb, TCA_FLOWER_MAX, tca[TCA_OPTIONS], + fl_policy, NULL); + if (err) + goto errout_tb; + + tmplt = kzalloc(sizeof(*tmplt), GFP_KERNEL); + if (!tmplt) { + err = -ENOMEM; + goto errout_tb; + } + tmplt->chain = chain; + err = fl_set_key(net, tb, &tmplt->dummy_key, &tmplt->mask, extack); + if (err) + goto errout_tmplt; + kfree(tb); + + fl_init_dissector(&tmplt->dissector, &tmplt->mask); + + fl_hw_create_tmplt(chain, tmplt); + + return tmplt; + +errout_tmplt: + kfree(tmplt); +errout_tb: + kfree(tb); + return ERR_PTR(err); +} + +static void fl_tmplt_destroy(void *tmplt_priv) +{ + struct fl_flow_tmplt *tmplt = tmplt_priv; + + fl_hw_destroy_tmplt(tmplt->chain, tmplt); + kfree(tmplt); +} + +static int fl_dump_key_val(struct sk_buff *skb, + void *val, int val_type, + void *mask, int mask_type, int len) +{ + int err; + + if (!memchr_inv(mask, 0, len)) + return 0; + err = nla_put(skb, val_type, len, val); + if (err) + return err; + if (mask_type != TCA_FLOWER_UNSPEC) { + err = nla_put(skb, mask_type, len, mask); + if (err) + return err; + } + return 0; +} + +static int fl_dump_key_mpls(struct sk_buff *skb, + struct flow_dissector_key_mpls *mpls_key, + struct flow_dissector_key_mpls *mpls_mask) +{ + int err; + + if (!memchr_inv(mpls_mask, 0, sizeof(*mpls_mask))) + return 0; + if (mpls_mask->mpls_ttl) { + err = nla_put_u8(skb, TCA_FLOWER_KEY_MPLS_TTL, + mpls_key->mpls_ttl); + if (err) + return err; + } + if (mpls_mask->mpls_tc) { + err = nla_put_u8(skb, TCA_FLOWER_KEY_MPLS_TC, + mpls_key->mpls_tc); + if (err) + return err; + } + if (mpls_mask->mpls_label) { + err = nla_put_u32(skb, TCA_FLOWER_KEY_MPLS_LABEL, + mpls_key->mpls_label); + if (err) + return err; + } + if (mpls_mask->mpls_bos) { + err = nla_put_u8(skb, TCA_FLOWER_KEY_MPLS_BOS, + mpls_key->mpls_bos); + if (err) + return err; + } + return 0; +} + +static int fl_dump_key_ip(struct sk_buff *skb, bool encap, + struct flow_dissector_key_ip *key, + struct flow_dissector_key_ip *mask) +{ + int tos_key = encap ? TCA_FLOWER_KEY_ENC_IP_TOS : TCA_FLOWER_KEY_IP_TOS; + int ttl_key = encap ? TCA_FLOWER_KEY_ENC_IP_TTL : TCA_FLOWER_KEY_IP_TTL; + int tos_mask = encap ? TCA_FLOWER_KEY_ENC_IP_TOS_MASK : TCA_FLOWER_KEY_IP_TOS_MASK; + int ttl_mask = encap ? TCA_FLOWER_KEY_ENC_IP_TTL_MASK : TCA_FLOWER_KEY_IP_TTL_MASK; + + if (fl_dump_key_val(skb, &key->tos, tos_key, &mask->tos, tos_mask, sizeof(key->tos)) || + fl_dump_key_val(skb, &key->ttl, ttl_key, &mask->ttl, ttl_mask, sizeof(key->ttl))) + return -1; + + return 0; +} + +static int fl_dump_key_vlan(struct sk_buff *skb, + int vlan_id_key, int vlan_prio_key, + struct flow_dissector_key_vlan *vlan_key, + struct flow_dissector_key_vlan *vlan_mask) +{ + int err; + + if (!memchr_inv(vlan_mask, 0, sizeof(*vlan_mask))) + return 0; + if (vlan_mask->vlan_id) { + err = nla_put_u16(skb, vlan_id_key, + vlan_key->vlan_id); + if (err) + return err; + } + if (vlan_mask->vlan_priority) { + err = nla_put_u8(skb, vlan_prio_key, + vlan_key->vlan_priority); + if (err) + return err; + } + return 0; +} + +static void fl_get_key_flag(u32 dissector_key, u32 dissector_mask, + u32 *flower_key, u32 *flower_mask, + u32 flower_flag_bit, u32 dissector_flag_bit) +{ + if (dissector_mask & dissector_flag_bit) { + *flower_mask |= flower_flag_bit; + if (dissector_key & dissector_flag_bit) + *flower_key |= flower_flag_bit; + } +} + +static int fl_dump_key_flags(struct sk_buff *skb, u32 flags_key, u32 flags_mask) +{ + u32 key, mask; + __be32 _key, _mask; + int err; + + if (!memchr_inv(&flags_mask, 0, sizeof(flags_mask))) + return 0; + + key = 0; + mask = 0; + + fl_get_key_flag(flags_key, flags_mask, &key, &mask, + TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT, FLOW_DIS_IS_FRAGMENT); + fl_get_key_flag(flags_key, flags_mask, &key, &mask, + TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST, + FLOW_DIS_FIRST_FRAG); + + _key = cpu_to_be32(key); + _mask = cpu_to_be32(mask); + + err = nla_put(skb, TCA_FLOWER_KEY_FLAGS, 4, &_key); + if (err) + return err; + + return nla_put(skb, TCA_FLOWER_KEY_FLAGS_MASK, 4, &_mask); +} + +static int fl_dump_key_geneve_opt(struct sk_buff *skb, + struct flow_dissector_key_enc_opts *enc_opts) +{ + struct geneve_opt *opt; + struct nlattr *nest; + int opt_off = 0; + + nest = nla_nest_start(skb, TCA_FLOWER_KEY_ENC_OPTS_GENEVE); + if (!nest) + goto nla_put_failure; + + while (enc_opts->len > opt_off) { + opt = (struct geneve_opt *)&enc_opts->data[opt_off]; + + if (nla_put_be16(skb, TCA_FLOWER_KEY_ENC_OPT_GENEVE_CLASS, + opt->opt_class)) + goto nla_put_failure; + if (nla_put_u8(skb, TCA_FLOWER_KEY_ENC_OPT_GENEVE_TYPE, + opt->type)) + goto nla_put_failure; + if (nla_put(skb, TCA_FLOWER_KEY_ENC_OPT_GENEVE_DATA, + opt->length * 4, opt->opt_data)) + goto nla_put_failure; + + opt_off += sizeof(struct geneve_opt) + opt->length * 4; + } + nla_nest_end(skb, nest); + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static int fl_dump_key_options(struct sk_buff *skb, int enc_opt_type, + struct flow_dissector_key_enc_opts *enc_opts) +{ + struct nlattr *nest; + int err; + + if (!enc_opts->len) + return 0; + + nest = nla_nest_start(skb, enc_opt_type); + if (!nest) + goto nla_put_failure; + + switch (enc_opts->dst_opt_type) { + case TUNNEL_GENEVE_OPT: + err = fl_dump_key_geneve_opt(skb, enc_opts); + if (err) + goto nla_put_failure; + break; + default: + goto nla_put_failure; + } + nla_nest_end(skb, nest); + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static int fl_dump_key_enc_opt(struct sk_buff *skb, + struct flow_dissector_key_enc_opts *key_opts, + struct flow_dissector_key_enc_opts *msk_opts) +{ + int err; + + err = fl_dump_key_options(skb, TCA_FLOWER_KEY_ENC_OPTS, key_opts); + if (err) + return err; + + return fl_dump_key_options(skb, TCA_FLOWER_KEY_ENC_OPTS_MASK, msk_opts); +} + +static int fl_dump_key(struct sk_buff *skb, struct net *net, + struct fl_flow_key *key, struct fl_flow_key *mask) +{ + if (mask->indev_ifindex) { + struct net_device *dev; + + dev = __dev_get_by_index(net, key->indev_ifindex); + if (dev && nla_put_string(skb, TCA_FLOWER_INDEV, dev->name)) + goto nla_put_failure; + } + + if (fl_dump_key_val(skb, key->eth.dst, TCA_FLOWER_KEY_ETH_DST, + mask->eth.dst, TCA_FLOWER_KEY_ETH_DST_MASK, + sizeof(key->eth.dst)) || + fl_dump_key_val(skb, key->eth.src, TCA_FLOWER_KEY_ETH_SRC, + mask->eth.src, TCA_FLOWER_KEY_ETH_SRC_MASK, + sizeof(key->eth.src)) || + fl_dump_key_val(skb, &key->basic.n_proto, TCA_FLOWER_KEY_ETH_TYPE, + &mask->basic.n_proto, TCA_FLOWER_UNSPEC, + sizeof(key->basic.n_proto))) + goto nla_put_failure; + + if (fl_dump_key_mpls(skb, &key->mpls, &mask->mpls)) + goto nla_put_failure; + + if (fl_dump_key_vlan(skb, TCA_FLOWER_KEY_VLAN_ID, + TCA_FLOWER_KEY_VLAN_PRIO, &key->vlan, &mask->vlan)) + goto nla_put_failure; + + if (fl_dump_key_vlan(skb, TCA_FLOWER_KEY_CVLAN_ID, + TCA_FLOWER_KEY_CVLAN_PRIO, + &key->cvlan, &mask->cvlan) || + (mask->cvlan.vlan_tpid && + nla_put_be16(skb, TCA_FLOWER_KEY_VLAN_ETH_TYPE, + key->cvlan.vlan_tpid))) + goto nla_put_failure; + + if (mask->basic.n_proto) { + if (mask->cvlan.vlan_eth_type) { + if (nla_put_be16(skb, TCA_FLOWER_KEY_CVLAN_ETH_TYPE, + key->basic.n_proto)) + goto nla_put_failure; + } else if (mask->vlan.vlan_eth_type) { + if (nla_put_be16(skb, TCA_FLOWER_KEY_VLAN_ETH_TYPE, + key->vlan.vlan_eth_type)) + goto nla_put_failure; + } + } + + if ((key->basic.n_proto == htons(ETH_P_IP) || + key->basic.n_proto == htons(ETH_P_IPV6)) && + (fl_dump_key_val(skb, &key->basic.ip_proto, TCA_FLOWER_KEY_IP_PROTO, + &mask->basic.ip_proto, TCA_FLOWER_UNSPEC, + sizeof(key->basic.ip_proto)) || + fl_dump_key_ip(skb, false, &key->ip, &mask->ip))) + goto nla_put_failure; + + if (key->control.addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS && + (fl_dump_key_val(skb, &key->ipv4.src, TCA_FLOWER_KEY_IPV4_SRC, + &mask->ipv4.src, TCA_FLOWER_KEY_IPV4_SRC_MASK, + sizeof(key->ipv4.src)) || + fl_dump_key_val(skb, &key->ipv4.dst, TCA_FLOWER_KEY_IPV4_DST, + &mask->ipv4.dst, TCA_FLOWER_KEY_IPV4_DST_MASK, + sizeof(key->ipv4.dst)))) + goto nla_put_failure; + else if (key->control.addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS && + (fl_dump_key_val(skb, &key->ipv6.src, TCA_FLOWER_KEY_IPV6_SRC, + &mask->ipv6.src, TCA_FLOWER_KEY_IPV6_SRC_MASK, + sizeof(key->ipv6.src)) || + fl_dump_key_val(skb, &key->ipv6.dst, TCA_FLOWER_KEY_IPV6_DST, + &mask->ipv6.dst, TCA_FLOWER_KEY_IPV6_DST_MASK, + sizeof(key->ipv6.dst)))) + goto nla_put_failure; + + if (key->basic.ip_proto == IPPROTO_TCP && + (fl_dump_key_val(skb, &key->tp.src, TCA_FLOWER_KEY_TCP_SRC, + &mask->tp.src, TCA_FLOWER_KEY_TCP_SRC_MASK, + sizeof(key->tp.src)) || + fl_dump_key_val(skb, &key->tp.dst, TCA_FLOWER_KEY_TCP_DST, + &mask->tp.dst, TCA_FLOWER_KEY_TCP_DST_MASK, + sizeof(key->tp.dst)) || + fl_dump_key_val(skb, &key->tcp.flags, TCA_FLOWER_KEY_TCP_FLAGS, + &mask->tcp.flags, TCA_FLOWER_KEY_TCP_FLAGS_MASK, + sizeof(key->tcp.flags)))) + goto nla_put_failure; + else if (key->basic.ip_proto == IPPROTO_UDP && + (fl_dump_key_val(skb, &key->tp.src, TCA_FLOWER_KEY_UDP_SRC, + &mask->tp.src, TCA_FLOWER_KEY_UDP_SRC_MASK, + sizeof(key->tp.src)) || + fl_dump_key_val(skb, &key->tp.dst, TCA_FLOWER_KEY_UDP_DST, + &mask->tp.dst, TCA_FLOWER_KEY_UDP_DST_MASK, + sizeof(key->tp.dst)))) + goto nla_put_failure; + else if (key->basic.ip_proto == IPPROTO_SCTP && + (fl_dump_key_val(skb, &key->tp.src, TCA_FLOWER_KEY_SCTP_SRC, + &mask->tp.src, TCA_FLOWER_KEY_SCTP_SRC_MASK, + sizeof(key->tp.src)) || + fl_dump_key_val(skb, &key->tp.dst, TCA_FLOWER_KEY_SCTP_DST, + &mask->tp.dst, TCA_FLOWER_KEY_SCTP_DST_MASK, + sizeof(key->tp.dst)))) + goto nla_put_failure; + else if (key->basic.n_proto == htons(ETH_P_IP) && + key->basic.ip_proto == IPPROTO_ICMP && + (fl_dump_key_val(skb, &key->icmp.type, + TCA_FLOWER_KEY_ICMPV4_TYPE, &mask->icmp.type, + TCA_FLOWER_KEY_ICMPV4_TYPE_MASK, + sizeof(key->icmp.type)) || + fl_dump_key_val(skb, &key->icmp.code, + TCA_FLOWER_KEY_ICMPV4_CODE, &mask->icmp.code, + TCA_FLOWER_KEY_ICMPV4_CODE_MASK, + sizeof(key->icmp.code)))) + goto nla_put_failure; + else if (key->basic.n_proto == htons(ETH_P_IPV6) && + key->basic.ip_proto == IPPROTO_ICMPV6 && + (fl_dump_key_val(skb, &key->icmp.type, + TCA_FLOWER_KEY_ICMPV6_TYPE, &mask->icmp.type, + TCA_FLOWER_KEY_ICMPV6_TYPE_MASK, + sizeof(key->icmp.type)) || + fl_dump_key_val(skb, &key->icmp.code, + TCA_FLOWER_KEY_ICMPV6_CODE, &mask->icmp.code, + TCA_FLOWER_KEY_ICMPV6_CODE_MASK, + sizeof(key->icmp.code)))) + goto nla_put_failure; + else if ((key->basic.n_proto == htons(ETH_P_ARP) || + key->basic.n_proto == htons(ETH_P_RARP)) && + (fl_dump_key_val(skb, &key->arp.sip, + TCA_FLOWER_KEY_ARP_SIP, &mask->arp.sip, + TCA_FLOWER_KEY_ARP_SIP_MASK, + sizeof(key->arp.sip)) || + fl_dump_key_val(skb, &key->arp.tip, + TCA_FLOWER_KEY_ARP_TIP, &mask->arp.tip, + TCA_FLOWER_KEY_ARP_TIP_MASK, + sizeof(key->arp.tip)) || + fl_dump_key_val(skb, &key->arp.op, + TCA_FLOWER_KEY_ARP_OP, &mask->arp.op, + TCA_FLOWER_KEY_ARP_OP_MASK, + sizeof(key->arp.op)) || + fl_dump_key_val(skb, key->arp.sha, TCA_FLOWER_KEY_ARP_SHA, + mask->arp.sha, TCA_FLOWER_KEY_ARP_SHA_MASK, + sizeof(key->arp.sha)) || + fl_dump_key_val(skb, key->arp.tha, TCA_FLOWER_KEY_ARP_THA, + mask->arp.tha, TCA_FLOWER_KEY_ARP_THA_MASK, + sizeof(key->arp.tha)))) + goto nla_put_failure; + + if (key->enc_control.addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS && + (fl_dump_key_val(skb, &key->enc_ipv4.src, + TCA_FLOWER_KEY_ENC_IPV4_SRC, &mask->enc_ipv4.src, + TCA_FLOWER_KEY_ENC_IPV4_SRC_MASK, + sizeof(key->enc_ipv4.src)) || + fl_dump_key_val(skb, &key->enc_ipv4.dst, + TCA_FLOWER_KEY_ENC_IPV4_DST, &mask->enc_ipv4.dst, + TCA_FLOWER_KEY_ENC_IPV4_DST_MASK, + sizeof(key->enc_ipv4.dst)))) + goto nla_put_failure; + else if (key->enc_control.addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS && + (fl_dump_key_val(skb, &key->enc_ipv6.src, + TCA_FLOWER_KEY_ENC_IPV6_SRC, &mask->enc_ipv6.src, + TCA_FLOWER_KEY_ENC_IPV6_SRC_MASK, + sizeof(key->enc_ipv6.src)) || + fl_dump_key_val(skb, &key->enc_ipv6.dst, + TCA_FLOWER_KEY_ENC_IPV6_DST, + &mask->enc_ipv6.dst, + TCA_FLOWER_KEY_ENC_IPV6_DST_MASK, + sizeof(key->enc_ipv6.dst)))) + goto nla_put_failure; + + if (fl_dump_key_val(skb, &key->enc_key_id, TCA_FLOWER_KEY_ENC_KEY_ID, + &mask->enc_key_id, TCA_FLOWER_UNSPEC, + sizeof(key->enc_key_id)) || + fl_dump_key_val(skb, &key->enc_tp.src, + TCA_FLOWER_KEY_ENC_UDP_SRC_PORT, + &mask->enc_tp.src, + TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK, + sizeof(key->enc_tp.src)) || + fl_dump_key_val(skb, &key->enc_tp.dst, + TCA_FLOWER_KEY_ENC_UDP_DST_PORT, + &mask->enc_tp.dst, + TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK, + sizeof(key->enc_tp.dst)) || + fl_dump_key_ip(skb, true, &key->enc_ip, &mask->enc_ip) || + fl_dump_key_enc_opt(skb, &key->enc_opts, &mask->enc_opts)) + goto nla_put_failure; + + if (fl_dump_key_flags(skb, key->control.flags, mask->control.flags)) + goto nla_put_failure; + + return 0; + +nla_put_failure: + return -EMSGSIZE; +} + +static int fl_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct cls_fl_filter *f = fh; + struct nlattr *nest; + struct fl_flow_key *key, *mask; + + if (!f) + return skb->len; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + if (f->res.classid && + nla_put_u32(skb, TCA_FLOWER_CLASSID, f->res.classid)) + goto nla_put_failure; + + key = &f->key; + mask = &f->mask->key; + + if (fl_dump_key(skb, net, key, mask)) + goto nla_put_failure; + + if (!tc_skip_hw(f->flags)) + fl_hw_update_stats(tp, f); + + if (f->flags && nla_put_u32(skb, TCA_FLOWER_FLAGS, f->flags)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int fl_tmplt_dump(struct sk_buff *skb, struct net *net, void *tmplt_priv) +{ + struct fl_flow_tmplt *tmplt = tmplt_priv; + struct fl_flow_key *key, *mask; + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + key = &tmplt->dummy_key; + mask = &tmplt->mask; + + if (fl_dump_key(skb, net, key, mask)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static void fl_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct cls_fl_filter *f = fh; + + if (f && f->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &f->res, base); + else + __tcf_unbind_filter(q, &f->res); + } +} + +static struct tcf_proto_ops cls_fl_ops __read_mostly = { + .kind = "flower", + .classify = fl_classify, + .init = fl_init, + .destroy = fl_destroy, + .get = fl_get, + .change = fl_change, + .delete = fl_delete, + .walk = fl_walk, + .reoffload = fl_reoffload, + .dump = fl_dump, + .bind_class = fl_bind_class, + .tmplt_create = fl_tmplt_create, + .tmplt_destroy = fl_tmplt_destroy, + .tmplt_dump = fl_tmplt_dump, + .owner = THIS_MODULE, +}; + +static int __init cls_fl_init(void) +{ + return register_tcf_proto_ops(&cls_fl_ops); +} + +static void __exit cls_fl_exit(void) +{ + unregister_tcf_proto_ops(&cls_fl_ops); +} + +module_init(cls_fl_init); +module_exit(cls_fl_exit); + +MODULE_AUTHOR("Jiri Pirko <jiri@resnulli.us>"); +MODULE_DESCRIPTION("Flower classifier"); +MODULE_LICENSE("GPL v2"); diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c new file mode 100644 index 000000000..cb2c62605 --- /dev/null +++ b/net/sched/cls_fw.c @@ -0,0 +1,474 @@ +/* + * net/sched/cls_fw.c Classifier mapping ipchains' fwmark to traffic class. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + * Changes: + * Karlis Peisenieks <karlis@mt.lv> : 990415 : fw_walk off by one + * Karlis Peisenieks <karlis@mt.lv> : 990415 : fw_delete killed all the filter (and kernel). + * Alex <alex@pilotsoft.com> : 2004xxyy: Added Action extension + * + * JHS: We should remove the CONFIG_NET_CLS_IND from here + * eventually when the meta match extension is made available + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netlink.h> +#include <net/act_api.h> +#include <net/pkt_cls.h> +#include <net/sch_generic.h> + +#define HTSIZE 256 + +struct fw_head { + u32 mask; + struct fw_filter __rcu *ht[HTSIZE]; + struct rcu_head rcu; +}; + +struct fw_filter { + struct fw_filter __rcu *next; + u32 id; + struct tcf_result res; +#ifdef CONFIG_NET_CLS_IND + int ifindex; +#endif /* CONFIG_NET_CLS_IND */ + struct tcf_exts exts; + struct tcf_proto *tp; + struct rcu_work rwork; +}; + +static u32 fw_hash(u32 handle) +{ + handle ^= (handle >> 16); + handle ^= (handle >> 8); + return handle % HTSIZE; +} + +static int fw_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct fw_head *head = rcu_dereference_bh(tp->root); + struct fw_filter *f; + int r; + u32 id = skb->mark; + + if (head != NULL) { + id &= head->mask; + + for (f = rcu_dereference_bh(head->ht[fw_hash(id)]); f; + f = rcu_dereference_bh(f->next)) { + if (f->id == id) { + *res = f->res; +#ifdef CONFIG_NET_CLS_IND + if (!tcf_match_indev(skb, f->ifindex)) + continue; +#endif /* CONFIG_NET_CLS_IND */ + r = tcf_exts_exec(skb, &f->exts, res); + if (r < 0) + continue; + + return r; + } + } + } else { + struct Qdisc *q = tcf_block_q(tp->chain->block); + + /* Old method: classify the packet using its skb mark. */ + if (id && (TC_H_MAJ(id) == 0 || + !(TC_H_MAJ(id ^ q->handle)))) { + res->classid = id; + res->class = 0; + return 0; + } + } + + return -1; +} + +static void *fw_get(struct tcf_proto *tp, u32 handle) +{ + struct fw_head *head = rtnl_dereference(tp->root); + struct fw_filter *f; + + if (head == NULL) + return NULL; + + f = rtnl_dereference(head->ht[fw_hash(handle)]); + for (; f; f = rtnl_dereference(f->next)) { + if (f->id == handle) + return f; + } + return NULL; +} + +static int fw_init(struct tcf_proto *tp) +{ + /* We don't allocate fw_head here, because in the old method + * we don't need it at all. + */ + return 0; +} + +static void __fw_delete_filter(struct fw_filter *f) +{ + tcf_exts_destroy(&f->exts); + tcf_exts_put_net(&f->exts); + kfree(f); +} + +static void fw_delete_filter_work(struct work_struct *work) +{ + struct fw_filter *f = container_of(to_rcu_work(work), + struct fw_filter, + rwork); + rtnl_lock(); + __fw_delete_filter(f); + rtnl_unlock(); +} + +static void fw_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct fw_head *head = rtnl_dereference(tp->root); + struct fw_filter *f; + int h; + + if (head == NULL) + return; + + for (h = 0; h < HTSIZE; h++) { + while ((f = rtnl_dereference(head->ht[h])) != NULL) { + RCU_INIT_POINTER(head->ht[h], + rtnl_dereference(f->next)); + tcf_unbind_filter(tp, &f->res); + if (tcf_exts_get_net(&f->exts)) + tcf_queue_work(&f->rwork, fw_delete_filter_work); + else + __fw_delete_filter(f); + } + } + kfree_rcu(head, rcu); +} + +static int fw_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct fw_head *head = rtnl_dereference(tp->root); + struct fw_filter *f = arg; + struct fw_filter __rcu **fp; + struct fw_filter *pfp; + int ret = -EINVAL; + int h; + + if (head == NULL || f == NULL) + goto out; + + fp = &head->ht[fw_hash(f->id)]; + + for (pfp = rtnl_dereference(*fp); pfp; + fp = &pfp->next, pfp = rtnl_dereference(*fp)) { + if (pfp == f) { + RCU_INIT_POINTER(*fp, rtnl_dereference(f->next)); + tcf_unbind_filter(tp, &f->res); + tcf_exts_get_net(&f->exts); + tcf_queue_work(&f->rwork, fw_delete_filter_work); + ret = 0; + break; + } + } + + *last = true; + for (h = 0; h < HTSIZE; h++) { + if (rcu_access_pointer(head->ht[h])) { + *last = false; + break; + } + } + +out: + return ret; +} + +static const struct nla_policy fw_policy[TCA_FW_MAX + 1] = { + [TCA_FW_CLASSID] = { .type = NLA_U32 }, + [TCA_FW_INDEV] = { .type = NLA_STRING, .len = IFNAMSIZ }, + [TCA_FW_MASK] = { .type = NLA_U32 }, +}; + +static int fw_set_parms(struct net *net, struct tcf_proto *tp, + struct fw_filter *f, struct nlattr **tb, + struct nlattr **tca, unsigned long base, bool ovr, + struct netlink_ext_ack *extack) +{ + struct fw_head *head = rtnl_dereference(tp->root); + u32 mask; + int err; + + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &f->exts, ovr, + extack); + if (err < 0) + return err; + + if (tb[TCA_FW_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_FW_CLASSID]); + tcf_bind_filter(tp, &f->res, base); + } + +#ifdef CONFIG_NET_CLS_IND + if (tb[TCA_FW_INDEV]) { + int ret; + ret = tcf_change_indev(net, tb[TCA_FW_INDEV], extack); + if (ret < 0) + return ret; + f->ifindex = ret; + } +#endif /* CONFIG_NET_CLS_IND */ + + err = -EINVAL; + if (tb[TCA_FW_MASK]) { + mask = nla_get_u32(tb[TCA_FW_MASK]); + if (mask != head->mask) + return err; + } else if (head->mask != 0xFFFFFFFF) + return err; + + return 0; +} + +static int fw_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, void **arg, + bool ovr, struct netlink_ext_ack *extack) +{ + struct fw_head *head = rtnl_dereference(tp->root); + struct fw_filter *f = *arg; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_FW_MAX + 1]; + int err; + + if (!opt) + return handle ? -EINVAL : 0; /* Succeed if it is old method. */ + + err = nla_parse_nested(tb, TCA_FW_MAX, opt, fw_policy, NULL); + if (err < 0) + return err; + + if (f) { + struct fw_filter *pfp, *fnew; + struct fw_filter __rcu **fp; + + if (f->id != handle && handle) + return -EINVAL; + + fnew = kzalloc(sizeof(struct fw_filter), GFP_KERNEL); + if (!fnew) + return -ENOBUFS; + + fnew->id = f->id; + fnew->res = f->res; +#ifdef CONFIG_NET_CLS_IND + fnew->ifindex = f->ifindex; +#endif /* CONFIG_NET_CLS_IND */ + fnew->tp = f->tp; + + err = tcf_exts_init(&fnew->exts, TCA_FW_ACT, TCA_FW_POLICE); + if (err < 0) { + kfree(fnew); + return err; + } + + err = fw_set_parms(net, tp, fnew, tb, tca, base, ovr, extack); + if (err < 0) { + tcf_exts_destroy(&fnew->exts); + kfree(fnew); + return err; + } + + fp = &head->ht[fw_hash(fnew->id)]; + for (pfp = rtnl_dereference(*fp); pfp; + fp = &pfp->next, pfp = rtnl_dereference(*fp)) + if (pfp == f) + break; + + RCU_INIT_POINTER(fnew->next, rtnl_dereference(pfp->next)); + rcu_assign_pointer(*fp, fnew); + tcf_unbind_filter(tp, &f->res); + tcf_exts_get_net(&f->exts); + tcf_queue_work(&f->rwork, fw_delete_filter_work); + + *arg = fnew; + return err; + } + + if (!handle) + return -EINVAL; + + if (!head) { + u32 mask = 0xFFFFFFFF; + if (tb[TCA_FW_MASK]) + mask = nla_get_u32(tb[TCA_FW_MASK]); + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOBUFS; + head->mask = mask; + + rcu_assign_pointer(tp->root, head); + } + + f = kzalloc(sizeof(struct fw_filter), GFP_KERNEL); + if (f == NULL) + return -ENOBUFS; + + err = tcf_exts_init(&f->exts, TCA_FW_ACT, TCA_FW_POLICE); + if (err < 0) + goto errout; + f->id = handle; + f->tp = tp; + + err = fw_set_parms(net, tp, f, tb, tca, base, ovr, extack); + if (err < 0) + goto errout; + + RCU_INIT_POINTER(f->next, head->ht[fw_hash(handle)]); + rcu_assign_pointer(head->ht[fw_hash(handle)], f); + + *arg = f; + return 0; + +errout: + tcf_exts_destroy(&f->exts); + kfree(f); + return err; +} + +static void fw_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct fw_head *head = rtnl_dereference(tp->root); + int h; + + if (head == NULL) + arg->stop = 1; + + if (arg->stop) + return; + + for (h = 0; h < HTSIZE; h++) { + struct fw_filter *f; + + for (f = rtnl_dereference(head->ht[h]); f; + f = rtnl_dereference(f->next)) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(tp, f, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static int fw_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct fw_head *head = rtnl_dereference(tp->root); + struct fw_filter *f = fh; + struct nlattr *nest; + + if (f == NULL) + return skb->len; + + t->tcm_handle = f->id; + + if (!f->res.classid && !tcf_exts_has_actions(&f->exts)) + return skb->len; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (f->res.classid && + nla_put_u32(skb, TCA_FW_CLASSID, f->res.classid)) + goto nla_put_failure; +#ifdef CONFIG_NET_CLS_IND + if (f->ifindex) { + struct net_device *dev; + dev = __dev_get_by_index(net, f->ifindex); + if (dev && nla_put_string(skb, TCA_FW_INDEV, dev->name)) + goto nla_put_failure; + } +#endif /* CONFIG_NET_CLS_IND */ + if (head->mask != 0xFFFFFFFF && + nla_put_u32(skb, TCA_FW_MASK, head->mask)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts) < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void fw_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct fw_filter *f = fh; + + if (f && f->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &f->res, base); + else + __tcf_unbind_filter(q, &f->res); + } +} + +static struct tcf_proto_ops cls_fw_ops __read_mostly = { + .kind = "fw", + .classify = fw_classify, + .init = fw_init, + .destroy = fw_destroy, + .get = fw_get, + .change = fw_change, + .delete = fw_delete, + .walk = fw_walk, + .dump = fw_dump, + .bind_class = fw_bind_class, + .owner = THIS_MODULE, +}; + +static int __init init_fw(void) +{ + return register_tcf_proto_ops(&cls_fw_ops); +} + +static void __exit exit_fw(void) +{ + unregister_tcf_proto_ops(&cls_fw_ops); +} + +module_init(init_fw) +module_exit(exit_fw) +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c new file mode 100644 index 000000000..74863b0ff --- /dev/null +++ b/net/sched/cls_matchall.c @@ -0,0 +1,357 @@ +/* + * net/sched/cls_matchll.c Match-all classifier + * + * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> + +#include <net/sch_generic.h> +#include <net/pkt_cls.h> + +struct cls_mall_head { + struct tcf_exts exts; + struct tcf_result res; + u32 handle; + u32 flags; + unsigned int in_hw_count; + struct rcu_work rwork; +}; + +static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct cls_mall_head *head = rcu_dereference_bh(tp->root); + + if (tc_skip_sw(head->flags)) + return -1; + + *res = head->res; + return tcf_exts_exec(skb, &head->exts, res); +} + +static int mall_init(struct tcf_proto *tp) +{ + return 0; +} + +static void __mall_destroy(struct cls_mall_head *head) +{ + tcf_exts_destroy(&head->exts); + tcf_exts_put_net(&head->exts); + kfree(head); +} + +static void mall_destroy_work(struct work_struct *work) +{ + struct cls_mall_head *head = container_of(to_rcu_work(work), + struct cls_mall_head, + rwork); + rtnl_lock(); + __mall_destroy(head); + rtnl_unlock(); +} + +static void mall_destroy_hw_filter(struct tcf_proto *tp, + struct cls_mall_head *head, + unsigned long cookie, + struct netlink_ext_ack *extack) +{ + struct tc_cls_matchall_offload cls_mall = {}; + struct tcf_block *block = tp->chain->block; + + tc_cls_common_offload_init(&cls_mall.common, tp, head->flags, extack); + cls_mall.command = TC_CLSMATCHALL_DESTROY; + cls_mall.cookie = cookie; + + tc_setup_cb_call(block, NULL, TC_SETUP_CLSMATCHALL, &cls_mall, false); + tcf_block_offload_dec(block, &head->flags); +} + +static int mall_replace_hw_filter(struct tcf_proto *tp, + struct cls_mall_head *head, + unsigned long cookie, + struct netlink_ext_ack *extack) +{ + struct tc_cls_matchall_offload cls_mall = {}; + struct tcf_block *block = tp->chain->block; + bool skip_sw = tc_skip_sw(head->flags); + int err; + + tc_cls_common_offload_init(&cls_mall.common, tp, head->flags, extack); + cls_mall.command = TC_CLSMATCHALL_REPLACE; + cls_mall.exts = &head->exts; + cls_mall.cookie = cookie; + + err = tc_setup_cb_call(block, NULL, TC_SETUP_CLSMATCHALL, + &cls_mall, skip_sw); + if (err < 0) { + mall_destroy_hw_filter(tp, head, cookie, NULL); + return err; + } else if (err > 0) { + head->in_hw_count = err; + tcf_block_offload_inc(block, &head->flags); + } + + if (skip_sw && !(head->flags & TCA_CLS_FLAGS_IN_HW)) + return -EINVAL; + + return 0; +} + +static void mall_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + + if (!head) + return; + + tcf_unbind_filter(tp, &head->res); + + if (!tc_skip_hw(head->flags)) + mall_destroy_hw_filter(tp, head, (unsigned long) head, extack); + + if (tcf_exts_get_net(&head->exts)) + tcf_queue_work(&head->rwork, mall_destroy_work); + else + __mall_destroy(head); +} + +static void *mall_get(struct tcf_proto *tp, u32 handle) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + + if (head && head->handle == handle) + return head; + + return NULL; +} + +static const struct nla_policy mall_policy[TCA_MATCHALL_MAX + 1] = { + [TCA_MATCHALL_UNSPEC] = { .type = NLA_UNSPEC }, + [TCA_MATCHALL_CLASSID] = { .type = NLA_U32 }, + [TCA_MATCHALL_FLAGS] = { .type = NLA_U32 }, +}; + +static int mall_set_parms(struct net *net, struct tcf_proto *tp, + struct cls_mall_head *head, + unsigned long base, struct nlattr **tb, + struct nlattr *est, bool ovr, + struct netlink_ext_ack *extack) +{ + int err; + + err = tcf_exts_validate(net, tp, tb, est, &head->exts, ovr, extack); + if (err < 0) + return err; + + if (tb[TCA_MATCHALL_CLASSID]) { + head->res.classid = nla_get_u32(tb[TCA_MATCHALL_CLASSID]); + tcf_bind_filter(tp, &head->res, base); + } + return 0; +} + +static int mall_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + void **arg, bool ovr, struct netlink_ext_ack *extack) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + struct cls_mall_head *new; + u32 flags = 0; + int err; + + if (!tca[TCA_OPTIONS]) + return -EINVAL; + + if (head) + return -EEXIST; + + err = nla_parse_nested(tb, TCA_MATCHALL_MAX, tca[TCA_OPTIONS], + mall_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_MATCHALL_FLAGS]) { + flags = nla_get_u32(tb[TCA_MATCHALL_FLAGS]); + if (!tc_flags_valid(flags)) + return -EINVAL; + } + + new = kzalloc(sizeof(*new), GFP_KERNEL); + if (!new) + return -ENOBUFS; + + err = tcf_exts_init(&new->exts, TCA_MATCHALL_ACT, 0); + if (err) + goto err_exts_init; + + if (!handle) + handle = 1; + new->handle = handle; + new->flags = flags; + + err = mall_set_parms(net, tp, new, base, tb, tca[TCA_RATE], ovr, + extack); + if (err) + goto err_set_parms; + + if (!tc_skip_hw(new->flags)) { + err = mall_replace_hw_filter(tp, new, (unsigned long)new, + extack); + if (err) + goto err_replace_hw_filter; + } + + if (!tc_in_hw(new->flags)) + new->flags |= TCA_CLS_FLAGS_NOT_IN_HW; + + *arg = head; + rcu_assign_pointer(tp->root, new); + return 0; + +err_replace_hw_filter: +err_set_parms: + tcf_exts_destroy(&new->exts); +err_exts_init: + kfree(new); + return err; +} + +static int mall_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + return -EOPNOTSUPP; +} + +static void mall_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + + if (arg->count < arg->skip) + goto skip; + if (arg->fn(tp, head, arg) < 0) + arg->stop = 1; +skip: + arg->count++; +} + +static int mall_reoffload(struct tcf_proto *tp, bool add, tc_setup_cb_t *cb, + void *cb_priv, struct netlink_ext_ack *extack) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct tc_cls_matchall_offload cls_mall = {}; + struct tcf_block *block = tp->chain->block; + int err; + + if (tc_skip_hw(head->flags)) + return 0; + + tc_cls_common_offload_init(&cls_mall.common, tp, head->flags, extack); + cls_mall.command = add ? + TC_CLSMATCHALL_REPLACE : TC_CLSMATCHALL_DESTROY; + cls_mall.exts = &head->exts; + cls_mall.cookie = (unsigned long)head; + + err = cb(TC_SETUP_CLSMATCHALL, &cls_mall, cb_priv); + if (err) { + if (add && tc_skip_sw(head->flags)) + return err; + return 0; + } + + tc_cls_offload_cnt_update(block, &head->in_hw_count, &head->flags, add); + + return 0; +} + +static int mall_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct cls_mall_head *head = fh; + struct nlattr *nest; + + if (!head) + return skb->len; + + t->tcm_handle = head->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + if (head->res.classid && + nla_put_u32(skb, TCA_MATCHALL_CLASSID, head->res.classid)) + goto nla_put_failure; + + if (head->flags && nla_put_u32(skb, TCA_MATCHALL_FLAGS, head->flags)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &head->exts)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &head->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void mall_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct cls_mall_head *head = fh; + + if (head && head->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &head->res, base); + else + __tcf_unbind_filter(q, &head->res); + } +} + +static struct tcf_proto_ops cls_mall_ops __read_mostly = { + .kind = "matchall", + .classify = mall_classify, + .init = mall_init, + .destroy = mall_destroy, + .get = mall_get, + .change = mall_change, + .delete = mall_delete, + .walk = mall_walk, + .reoffload = mall_reoffload, + .dump = mall_dump, + .bind_class = mall_bind_class, + .owner = THIS_MODULE, +}; + +static int __init cls_mall_init(void) +{ + return register_tcf_proto_ops(&cls_mall_ops); +} + +static void __exit cls_mall_exit(void) +{ + unregister_tcf_proto_ops(&cls_mall_ops); +} + +module_init(cls_mall_init); +module_exit(cls_mall_exit); + +MODULE_AUTHOR("Jiri Pirko <jiri@mellanox.com>"); +MODULE_DESCRIPTION("Match-all classifier"); +MODULE_LICENSE("GPL v2"); diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c new file mode 100644 index 000000000..0256777b8 --- /dev/null +++ b/net/sched/cls_route.c @@ -0,0 +1,687 @@ +/* + * net/sched/cls_route.c ROUTE4 classifier. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/dst.h> +#include <net/route.h> +#include <net/netlink.h> +#include <net/act_api.h> +#include <net/pkt_cls.h> + +/* + * 1. For now we assume that route tags < 256. + * It allows to use direct table lookups, instead of hash tables. + * 2. For now we assume that "from TAG" and "fromdev DEV" statements + * are mutually exclusive. + * 3. "to TAG from ANY" has higher priority, than "to ANY from XXX" + */ +struct route4_fastmap { + struct route4_filter *filter; + u32 id; + int iif; +}; + +struct route4_head { + struct route4_fastmap fastmap[16]; + struct route4_bucket __rcu *table[256 + 1]; + struct rcu_head rcu; +}; + +struct route4_bucket { + /* 16 FROM buckets + 16 IIF buckets + 1 wildcard bucket */ + struct route4_filter __rcu *ht[16 + 16 + 1]; + struct rcu_head rcu; +}; + +struct route4_filter { + struct route4_filter __rcu *next; + u32 id; + int iif; + + struct tcf_result res; + struct tcf_exts exts; + u32 handle; + struct route4_bucket *bkt; + struct tcf_proto *tp; + struct rcu_work rwork; +}; + +#define ROUTE4_FAILURE ((struct route4_filter *)(-1L)) + +static inline int route4_fastmap_hash(u32 id, int iif) +{ + return id & 0xF; +} + +static DEFINE_SPINLOCK(fastmap_lock); +static void +route4_reset_fastmap(struct route4_head *head) +{ + spin_lock_bh(&fastmap_lock); + memset(head->fastmap, 0, sizeof(head->fastmap)); + spin_unlock_bh(&fastmap_lock); +} + +static void +route4_set_fastmap(struct route4_head *head, u32 id, int iif, + struct route4_filter *f) +{ + int h = route4_fastmap_hash(id, iif); + + /* fastmap updates must look atomic to aling id, iff, filter */ + spin_lock_bh(&fastmap_lock); + head->fastmap[h].id = id; + head->fastmap[h].iif = iif; + head->fastmap[h].filter = f; + spin_unlock_bh(&fastmap_lock); +} + +static inline int route4_hash_to(u32 id) +{ + return id & 0xFF; +} + +static inline int route4_hash_from(u32 id) +{ + return (id >> 16) & 0xF; +} + +static inline int route4_hash_iif(int iif) +{ + return 16 + ((iif >> 16) & 0xF); +} + +static inline int route4_hash_wild(void) +{ + return 32; +} + +#define ROUTE4_APPLY_RESULT() \ +{ \ + *res = f->res; \ + if (tcf_exts_has_actions(&f->exts)) { \ + int r = tcf_exts_exec(skb, &f->exts, res); \ + if (r < 0) { \ + dont_cache = 1; \ + continue; \ + } \ + return r; \ + } else if (!dont_cache) \ + route4_set_fastmap(head, id, iif, f); \ + return 0; \ +} + +static int route4_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct route4_head *head = rcu_dereference_bh(tp->root); + struct dst_entry *dst; + struct route4_bucket *b; + struct route4_filter *f; + u32 id, h; + int iif, dont_cache = 0; + + dst = skb_dst(skb); + if (!dst) + goto failure; + + id = dst->tclassid; + + iif = inet_iif(skb); + + h = route4_fastmap_hash(id, iif); + + spin_lock(&fastmap_lock); + if (id == head->fastmap[h].id && + iif == head->fastmap[h].iif && + (f = head->fastmap[h].filter) != NULL) { + if (f == ROUTE4_FAILURE) { + spin_unlock(&fastmap_lock); + goto failure; + } + + *res = f->res; + spin_unlock(&fastmap_lock); + return 0; + } + spin_unlock(&fastmap_lock); + + h = route4_hash_to(id); + +restart: + b = rcu_dereference_bh(head->table[h]); + if (b) { + for (f = rcu_dereference_bh(b->ht[route4_hash_from(id)]); + f; + f = rcu_dereference_bh(f->next)) + if (f->id == id) + ROUTE4_APPLY_RESULT(); + + for (f = rcu_dereference_bh(b->ht[route4_hash_iif(iif)]); + f; + f = rcu_dereference_bh(f->next)) + if (f->iif == iif) + ROUTE4_APPLY_RESULT(); + + for (f = rcu_dereference_bh(b->ht[route4_hash_wild()]); + f; + f = rcu_dereference_bh(f->next)) + ROUTE4_APPLY_RESULT(); + } + if (h < 256) { + h = 256; + id &= ~0xFFFF; + goto restart; + } + + if (!dont_cache) + route4_set_fastmap(head, id, iif, ROUTE4_FAILURE); +failure: + return -1; +} + +static inline u32 to_hash(u32 id) +{ + u32 h = id & 0xFF; + + if (id & 0x8000) + h += 256; + return h; +} + +static inline u32 from_hash(u32 id) +{ + id &= 0xFFFF; + if (id == 0xFFFF) + return 32; + if (!(id & 0x8000)) { + if (id > 255) + return 256; + return id & 0xF; + } + return 16 + (id & 0xF); +} + +static void *route4_get(struct tcf_proto *tp, u32 handle) +{ + struct route4_head *head = rtnl_dereference(tp->root); + struct route4_bucket *b; + struct route4_filter *f; + unsigned int h1, h2; + + h1 = to_hash(handle); + if (h1 > 256) + return NULL; + + h2 = from_hash(handle >> 16); + if (h2 > 32) + return NULL; + + b = rtnl_dereference(head->table[h1]); + if (b) { + for (f = rtnl_dereference(b->ht[h2]); + f; + f = rtnl_dereference(f->next)) + if (f->handle == handle) + return f; + } + return NULL; +} + +static int route4_init(struct tcf_proto *tp) +{ + struct route4_head *head; + + head = kzalloc(sizeof(struct route4_head), GFP_KERNEL); + if (head == NULL) + return -ENOBUFS; + + rcu_assign_pointer(tp->root, head); + return 0; +} + +static void __route4_delete_filter(struct route4_filter *f) +{ + tcf_exts_destroy(&f->exts); + tcf_exts_put_net(&f->exts); + kfree(f); +} + +static void route4_delete_filter_work(struct work_struct *work) +{ + struct route4_filter *f = container_of(to_rcu_work(work), + struct route4_filter, + rwork); + rtnl_lock(); + __route4_delete_filter(f); + rtnl_unlock(); +} + +static void route4_queue_work(struct route4_filter *f) +{ + tcf_queue_work(&f->rwork, route4_delete_filter_work); +} + +static void route4_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct route4_head *head = rtnl_dereference(tp->root); + int h1, h2; + + if (head == NULL) + return; + + for (h1 = 0; h1 <= 256; h1++) { + struct route4_bucket *b; + + b = rtnl_dereference(head->table[h1]); + if (b) { + for (h2 = 0; h2 <= 32; h2++) { + struct route4_filter *f; + + while ((f = rtnl_dereference(b->ht[h2])) != NULL) { + struct route4_filter *next; + + next = rtnl_dereference(f->next); + RCU_INIT_POINTER(b->ht[h2], next); + tcf_unbind_filter(tp, &f->res); + if (tcf_exts_get_net(&f->exts)) + route4_queue_work(f); + else + __route4_delete_filter(f); + } + } + RCU_INIT_POINTER(head->table[h1], NULL); + kfree_rcu(b, rcu); + } + } + kfree_rcu(head, rcu); +} + +static int route4_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct route4_head *head = rtnl_dereference(tp->root); + struct route4_filter *f = arg; + struct route4_filter __rcu **fp; + struct route4_filter *nf; + struct route4_bucket *b; + unsigned int h = 0; + int i, h1; + + if (!head || !f) + return -EINVAL; + + h = f->handle; + b = f->bkt; + + fp = &b->ht[from_hash(h >> 16)]; + for (nf = rtnl_dereference(*fp); nf; + fp = &nf->next, nf = rtnl_dereference(*fp)) { + if (nf == f) { + /* unlink it */ + RCU_INIT_POINTER(*fp, rtnl_dereference(f->next)); + + /* Remove any fastmap lookups that might ref filter + * notice we unlink'd the filter so we can't get it + * back in the fastmap. + */ + route4_reset_fastmap(head); + + /* Delete it */ + tcf_unbind_filter(tp, &f->res); + tcf_exts_get_net(&f->exts); + tcf_queue_work(&f->rwork, route4_delete_filter_work); + + /* Strip RTNL protected tree */ + for (i = 0; i <= 32; i++) { + struct route4_filter *rt; + + rt = rtnl_dereference(b->ht[i]); + if (rt) + goto out; + } + + /* OK, session has no flows */ + RCU_INIT_POINTER(head->table[to_hash(h)], NULL); + kfree_rcu(b, rcu); + break; + } + } + +out: + *last = true; + for (h1 = 0; h1 <= 256; h1++) { + if (rcu_access_pointer(head->table[h1])) { + *last = false; + break; + } + } + + return 0; +} + +static const struct nla_policy route4_policy[TCA_ROUTE4_MAX + 1] = { + [TCA_ROUTE4_CLASSID] = { .type = NLA_U32 }, + [TCA_ROUTE4_TO] = { .type = NLA_U32 }, + [TCA_ROUTE4_FROM] = { .type = NLA_U32 }, + [TCA_ROUTE4_IIF] = { .type = NLA_U32 }, +}; + +static int route4_set_parms(struct net *net, struct tcf_proto *tp, + unsigned long base, struct route4_filter *f, + u32 handle, struct route4_head *head, + struct nlattr **tb, struct nlattr *est, int new, + bool ovr, struct netlink_ext_ack *extack) +{ + u32 id = 0, to = 0, nhandle = 0x8000; + struct route4_filter *fp; + unsigned int h1; + struct route4_bucket *b; + int err; + + err = tcf_exts_validate(net, tp, tb, est, &f->exts, ovr, extack); + if (err < 0) + return err; + + if (tb[TCA_ROUTE4_TO]) { + if (new && handle & 0x8000) + return -EINVAL; + to = nla_get_u32(tb[TCA_ROUTE4_TO]); + if (to > 0xFF) + return -EINVAL; + nhandle = to; + } + + if (tb[TCA_ROUTE4_FROM]) { + if (tb[TCA_ROUTE4_IIF]) + return -EINVAL; + id = nla_get_u32(tb[TCA_ROUTE4_FROM]); + if (id > 0xFF) + return -EINVAL; + nhandle |= id << 16; + } else if (tb[TCA_ROUTE4_IIF]) { + id = nla_get_u32(tb[TCA_ROUTE4_IIF]); + if (id > 0x7FFF) + return -EINVAL; + nhandle |= (id | 0x8000) << 16; + } else + nhandle |= 0xFFFF << 16; + + if (handle && new) { + nhandle |= handle & 0x7F00; + if (nhandle != handle) + return -EINVAL; + } + + h1 = to_hash(nhandle); + b = rtnl_dereference(head->table[h1]); + if (!b) { + b = kzalloc(sizeof(struct route4_bucket), GFP_KERNEL); + if (b == NULL) + return -ENOBUFS; + + rcu_assign_pointer(head->table[h1], b); + } else { + unsigned int h2 = from_hash(nhandle >> 16); + + for (fp = rtnl_dereference(b->ht[h2]); + fp; + fp = rtnl_dereference(fp->next)) + if (fp->handle == f->handle) + return -EEXIST; + } + + if (tb[TCA_ROUTE4_TO]) + f->id = to; + + if (tb[TCA_ROUTE4_FROM]) + f->id = to | id<<16; + else if (tb[TCA_ROUTE4_IIF]) + f->iif = id; + + f->handle = nhandle; + f->bkt = b; + f->tp = tp; + + if (tb[TCA_ROUTE4_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_ROUTE4_CLASSID]); + tcf_bind_filter(tp, &f->res, base); + } + + return 0; +} + +static int route4_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, u32 handle, + struct nlattr **tca, void **arg, bool ovr, + struct netlink_ext_ack *extack) +{ + struct route4_head *head = rtnl_dereference(tp->root); + struct route4_filter __rcu **fp; + struct route4_filter *fold, *f1, *pfp, *f = NULL; + struct route4_bucket *b; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_ROUTE4_MAX + 1]; + unsigned int h, th; + int err; + bool new = true; + + if (opt == NULL) + return handle ? -EINVAL : 0; + + err = nla_parse_nested(tb, TCA_ROUTE4_MAX, opt, route4_policy, NULL); + if (err < 0) + return err; + + fold = *arg; + if (fold && handle && fold->handle != handle) + return -EINVAL; + + err = -ENOBUFS; + f = kzalloc(sizeof(struct route4_filter), GFP_KERNEL); + if (!f) + goto errout; + + err = tcf_exts_init(&f->exts, TCA_ROUTE4_ACT, TCA_ROUTE4_POLICE); + if (err < 0) + goto errout; + + if (fold) { + f->id = fold->id; + f->iif = fold->iif; + f->res = fold->res; + f->handle = fold->handle; + + f->tp = fold->tp; + f->bkt = fold->bkt; + new = false; + } + + err = route4_set_parms(net, tp, base, f, handle, head, tb, + tca[TCA_RATE], new, ovr, extack); + if (err < 0) + goto errout; + + h = from_hash(f->handle >> 16); + fp = &f->bkt->ht[h]; + for (pfp = rtnl_dereference(*fp); + (f1 = rtnl_dereference(*fp)) != NULL; + fp = &f1->next) + if (f->handle < f1->handle) + break; + + tcf_block_netif_keep_dst(tp->chain->block); + rcu_assign_pointer(f->next, f1); + rcu_assign_pointer(*fp, f); + + if (fold && fold->handle && f->handle != fold->handle) { + th = to_hash(fold->handle); + h = from_hash(fold->handle >> 16); + b = rtnl_dereference(head->table[th]); + if (b) { + fp = &b->ht[h]; + for (pfp = rtnl_dereference(*fp); pfp; + fp = &pfp->next, pfp = rtnl_dereference(*fp)) { + if (pfp == fold) { + rcu_assign_pointer(*fp, fold->next); + break; + } + } + } + } + + route4_reset_fastmap(head); + *arg = f; + if (fold) { + tcf_unbind_filter(tp, &fold->res); + tcf_exts_get_net(&fold->exts); + tcf_queue_work(&fold->rwork, route4_delete_filter_work); + } + return 0; + +errout: + if (f) + tcf_exts_destroy(&f->exts); + kfree(f); + return err; +} + +static void route4_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct route4_head *head = rtnl_dereference(tp->root); + unsigned int h, h1; + + if (head == NULL) + arg->stop = 1; + + if (arg->stop) + return; + + for (h = 0; h <= 256; h++) { + struct route4_bucket *b = rtnl_dereference(head->table[h]); + + if (b) { + for (h1 = 0; h1 <= 32; h1++) { + struct route4_filter *f; + + for (f = rtnl_dereference(b->ht[h1]); + f; + f = rtnl_dereference(f->next)) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(tp, f, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } + } + } +} + +static int route4_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct route4_filter *f = fh; + struct nlattr *nest; + u32 id; + + if (f == NULL) + return skb->len; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (!(f->handle & 0x8000)) { + id = f->id & 0xFF; + if (nla_put_u32(skb, TCA_ROUTE4_TO, id)) + goto nla_put_failure; + } + if (f->handle & 0x80000000) { + if ((f->handle >> 16) != 0xFFFF && + nla_put_u32(skb, TCA_ROUTE4_IIF, f->iif)) + goto nla_put_failure; + } else { + id = f->id >> 16; + if (nla_put_u32(skb, TCA_ROUTE4_FROM, id)) + goto nla_put_failure; + } + if (f->res.classid && + nla_put_u32(skb, TCA_ROUTE4_CLASSID, f->res.classid)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts) < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void route4_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct route4_filter *f = fh; + + if (f && f->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &f->res, base); + else + __tcf_unbind_filter(q, &f->res); + } +} + +static struct tcf_proto_ops cls_route4_ops __read_mostly = { + .kind = "route", + .classify = route4_classify, + .init = route4_init, + .destroy = route4_destroy, + .get = route4_get, + .change = route4_change, + .delete = route4_delete, + .walk = route4_walk, + .dump = route4_dump, + .bind_class = route4_bind_class, + .owner = THIS_MODULE, +}; + +static int __init init_route4(void) +{ + return register_tcf_proto_ops(&cls_route4_ops); +} + +static void __exit exit_route4(void) +{ + unregister_tcf_proto_ops(&cls_route4_ops); +} + +module_init(init_route4) +module_exit(exit_route4) +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_rsvp.c b/net/sched/cls_rsvp.c new file mode 100644 index 000000000..cbb5e0d60 --- /dev/null +++ b/net/sched/cls_rsvp.c @@ -0,0 +1,28 @@ +/* + * net/sched/cls_rsvp.c Special RSVP packet classifier for IPv4. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/ip.h> +#include <net/netlink.h> +#include <net/act_api.h> +#include <net/pkt_cls.h> + +#define RSVP_DST_LEN 1 +#define RSVP_ID "rsvp" +#define RSVP_OPS cls_rsvp_ops + +#include "cls_rsvp.h" +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h new file mode 100644 index 000000000..eb1dd2afc --- /dev/null +++ b/net/sched/cls_rsvp.h @@ -0,0 +1,775 @@ +/* + * net/sched/cls_rsvp.h Template file for RSVPv[46] classifiers. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +/* + Comparing to general packet classification problem, + RSVP needs only sevaral relatively simple rules: + + * (dst, protocol) are always specified, + so that we are able to hash them. + * src may be exact, or may be wildcard, so that + we can keep a hash table plus one wildcard entry. + * source port (or flow label) is important only if src is given. + + IMPLEMENTATION. + + We use a two level hash table: The top level is keyed by + destination address and protocol ID, every bucket contains a list + of "rsvp sessions", identified by destination address, protocol and + DPI(="Destination Port ID"): triple (key, mask, offset). + + Every bucket has a smaller hash table keyed by source address + (cf. RSVP flowspec) and one wildcard entry for wildcard reservations. + Every bucket is again a list of "RSVP flows", selected by + source address and SPI(="Source Port ID" here rather than + "security parameter index"): triple (key, mask, offset). + + + NOTE 1. All the packets with IPv6 extension headers (but AH and ESP) + and all fragmented packets go to the best-effort traffic class. + + + NOTE 2. Two "port id"'s seems to be redundant, rfc2207 requires + only one "Generalized Port Identifier". So that for classic + ah, esp (and udp,tcp) both *pi should coincide or one of them + should be wildcard. + + At first sight, this redundancy is just a waste of CPU + resources. But DPI and SPI add the possibility to assign different + priorities to GPIs. Look also at note 4 about tunnels below. + + + NOTE 3. One complication is the case of tunneled packets. + We implement it as following: if the first lookup + matches a special session with "tunnelhdr" value not zero, + flowid doesn't contain the true flow ID, but the tunnel ID (1...255). + In this case, we pull tunnelhdr bytes and restart lookup + with tunnel ID added to the list of keys. Simple and stupid 8)8) + It's enough for PIMREG and IPIP. + + + NOTE 4. Two GPIs make it possible to parse even GRE packets. + F.e. DPI can select ETH_P_IP (and necessary flags to make + tunnelhdr correct) in GRE protocol field and SPI matches + GRE key. Is it not nice? 8)8) + + + Well, as result, despite its simplicity, we get a pretty + powerful classification engine. */ + + +struct rsvp_head { + u32 tmap[256/32]; + u32 hgenerator; + u8 tgenerator; + struct rsvp_session __rcu *ht[256]; + struct rcu_head rcu; +}; + +struct rsvp_session { + struct rsvp_session __rcu *next; + __be32 dst[RSVP_DST_LEN]; + struct tc_rsvp_gpi dpi; + u8 protocol; + u8 tunnelid; + /* 16 (src,sport) hash slots, and one wildcard source slot */ + struct rsvp_filter __rcu *ht[16 + 1]; + struct rcu_head rcu; +}; + + +struct rsvp_filter { + struct rsvp_filter __rcu *next; + __be32 src[RSVP_DST_LEN]; + struct tc_rsvp_gpi spi; + u8 tunnelhdr; + + struct tcf_result res; + struct tcf_exts exts; + + u32 handle; + struct rsvp_session *sess; + struct rcu_work rwork; +}; + +static inline unsigned int hash_dst(__be32 *dst, u8 protocol, u8 tunnelid) +{ + unsigned int h = (__force __u32)dst[RSVP_DST_LEN - 1]; + + h ^= h>>16; + h ^= h>>8; + return (h ^ protocol ^ tunnelid) & 0xFF; +} + +static inline unsigned int hash_src(__be32 *src) +{ + unsigned int h = (__force __u32)src[RSVP_DST_LEN-1]; + + h ^= h>>16; + h ^= h>>8; + h ^= h>>4; + return h & 0xF; +} + +#define RSVP_APPLY_RESULT() \ +{ \ + int r = tcf_exts_exec(skb, &f->exts, res); \ + if (r < 0) \ + continue; \ + else if (r > 0) \ + return r; \ +} + +static int rsvp_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct rsvp_head *head = rcu_dereference_bh(tp->root); + struct rsvp_session *s; + struct rsvp_filter *f; + unsigned int h1, h2; + __be32 *dst, *src; + u8 protocol; + u8 tunnelid = 0; + u8 *xprt; +#if RSVP_DST_LEN == 4 + struct ipv6hdr *nhptr; + + if (!pskb_network_may_pull(skb, sizeof(*nhptr))) + return -1; + nhptr = ipv6_hdr(skb); +#else + struct iphdr *nhptr; + + if (!pskb_network_may_pull(skb, sizeof(*nhptr))) + return -1; + nhptr = ip_hdr(skb); +#endif +restart: + +#if RSVP_DST_LEN == 4 + src = &nhptr->saddr.s6_addr32[0]; + dst = &nhptr->daddr.s6_addr32[0]; + protocol = nhptr->nexthdr; + xprt = ((u8 *)nhptr) + sizeof(struct ipv6hdr); +#else + src = &nhptr->saddr; + dst = &nhptr->daddr; + protocol = nhptr->protocol; + xprt = ((u8 *)nhptr) + (nhptr->ihl<<2); + if (ip_is_fragment(nhptr)) + return -1; +#endif + + h1 = hash_dst(dst, protocol, tunnelid); + h2 = hash_src(src); + + for (s = rcu_dereference_bh(head->ht[h1]); s; + s = rcu_dereference_bh(s->next)) { + if (dst[RSVP_DST_LEN-1] == s->dst[RSVP_DST_LEN - 1] && + protocol == s->protocol && + !(s->dpi.mask & + (*(u32 *)(xprt + s->dpi.offset) ^ s->dpi.key)) && +#if RSVP_DST_LEN == 4 + dst[0] == s->dst[0] && + dst[1] == s->dst[1] && + dst[2] == s->dst[2] && +#endif + tunnelid == s->tunnelid) { + + for (f = rcu_dereference_bh(s->ht[h2]); f; + f = rcu_dereference_bh(f->next)) { + if (src[RSVP_DST_LEN-1] == f->src[RSVP_DST_LEN - 1] && + !(f->spi.mask & (*(u32 *)(xprt + f->spi.offset) ^ f->spi.key)) +#if RSVP_DST_LEN == 4 + && + src[0] == f->src[0] && + src[1] == f->src[1] && + src[2] == f->src[2] +#endif + ) { + *res = f->res; + RSVP_APPLY_RESULT(); + +matched: + if (f->tunnelhdr == 0) + return 0; + + tunnelid = f->res.classid; + nhptr = (void *)(xprt + f->tunnelhdr - sizeof(*nhptr)); + goto restart; + } + } + + /* And wildcard bucket... */ + for (f = rcu_dereference_bh(s->ht[16]); f; + f = rcu_dereference_bh(f->next)) { + *res = f->res; + RSVP_APPLY_RESULT(); + goto matched; + } + return -1; + } + } + return -1; +} + +static void rsvp_replace(struct tcf_proto *tp, struct rsvp_filter *n, u32 h) +{ + struct rsvp_head *head = rtnl_dereference(tp->root); + struct rsvp_session *s; + struct rsvp_filter __rcu **ins; + struct rsvp_filter *pins; + unsigned int h1 = h & 0xFF; + unsigned int h2 = (h >> 8) & 0xFF; + + for (s = rtnl_dereference(head->ht[h1]); s; + s = rtnl_dereference(s->next)) { + for (ins = &s->ht[h2], pins = rtnl_dereference(*ins); ; + ins = &pins->next, pins = rtnl_dereference(*ins)) { + if (pins->handle == h) { + RCU_INIT_POINTER(n->next, pins->next); + rcu_assign_pointer(*ins, n); + return; + } + } + } + + /* Something went wrong if we are trying to replace a non-existant + * node. Mind as well halt instead of silently failing. + */ + BUG_ON(1); +} + +static void *rsvp_get(struct tcf_proto *tp, u32 handle) +{ + struct rsvp_head *head = rtnl_dereference(tp->root); + struct rsvp_session *s; + struct rsvp_filter *f; + unsigned int h1 = handle & 0xFF; + unsigned int h2 = (handle >> 8) & 0xFF; + + if (h2 > 16) + return NULL; + + for (s = rtnl_dereference(head->ht[h1]); s; + s = rtnl_dereference(s->next)) { + for (f = rtnl_dereference(s->ht[h2]); f; + f = rtnl_dereference(f->next)) { + if (f->handle == handle) + return f; + } + } + return NULL; +} + +static int rsvp_init(struct tcf_proto *tp) +{ + struct rsvp_head *data; + + data = kzalloc(sizeof(struct rsvp_head), GFP_KERNEL); + if (data) { + rcu_assign_pointer(tp->root, data); + return 0; + } + return -ENOBUFS; +} + +static void __rsvp_delete_filter(struct rsvp_filter *f) +{ + tcf_exts_destroy(&f->exts); + tcf_exts_put_net(&f->exts); + kfree(f); +} + +static void rsvp_delete_filter_work(struct work_struct *work) +{ + struct rsvp_filter *f = container_of(to_rcu_work(work), + struct rsvp_filter, + rwork); + rtnl_lock(); + __rsvp_delete_filter(f); + rtnl_unlock(); +} + +static void rsvp_delete_filter(struct tcf_proto *tp, struct rsvp_filter *f) +{ + tcf_unbind_filter(tp, &f->res); + /* all classifiers are required to call tcf_exts_destroy() after rcu + * grace period, since converted-to-rcu actions are relying on that + * in cleanup() callback + */ + if (tcf_exts_get_net(&f->exts)) + tcf_queue_work(&f->rwork, rsvp_delete_filter_work); + else + __rsvp_delete_filter(f); +} + +static void rsvp_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct rsvp_head *data = rtnl_dereference(tp->root); + int h1, h2; + + if (data == NULL) + return; + + for (h1 = 0; h1 < 256; h1++) { + struct rsvp_session *s; + + while ((s = rtnl_dereference(data->ht[h1])) != NULL) { + RCU_INIT_POINTER(data->ht[h1], s->next); + + for (h2 = 0; h2 <= 16; h2++) { + struct rsvp_filter *f; + + while ((f = rtnl_dereference(s->ht[h2])) != NULL) { + rcu_assign_pointer(s->ht[h2], f->next); + rsvp_delete_filter(tp, f); + } + } + kfree_rcu(s, rcu); + } + } + kfree_rcu(data, rcu); +} + +static int rsvp_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct rsvp_head *head = rtnl_dereference(tp->root); + struct rsvp_filter *nfp, *f = arg; + struct rsvp_filter __rcu **fp; + unsigned int h = f->handle; + struct rsvp_session __rcu **sp; + struct rsvp_session *nsp, *s = f->sess; + int i, h1; + + fp = &s->ht[(h >> 8) & 0xFF]; + for (nfp = rtnl_dereference(*fp); nfp; + fp = &nfp->next, nfp = rtnl_dereference(*fp)) { + if (nfp == f) { + RCU_INIT_POINTER(*fp, f->next); + rsvp_delete_filter(tp, f); + + /* Strip tree */ + + for (i = 0; i <= 16; i++) + if (s->ht[i]) + goto out; + + /* OK, session has no flows */ + sp = &head->ht[h & 0xFF]; + for (nsp = rtnl_dereference(*sp); nsp; + sp = &nsp->next, nsp = rtnl_dereference(*sp)) { + if (nsp == s) { + RCU_INIT_POINTER(*sp, s->next); + kfree_rcu(s, rcu); + goto out; + } + } + + break; + } + } + +out: + *last = true; + for (h1 = 0; h1 < 256; h1++) { + if (rcu_access_pointer(head->ht[h1])) { + *last = false; + break; + } + } + + return 0; +} + +static unsigned int gen_handle(struct tcf_proto *tp, unsigned salt) +{ + struct rsvp_head *data = rtnl_dereference(tp->root); + int i = 0xFFFF; + + while (i-- > 0) { + u32 h; + + if ((data->hgenerator += 0x10000) == 0) + data->hgenerator = 0x10000; + h = data->hgenerator|salt; + if (!rsvp_get(tp, h)) + return h; + } + return 0; +} + +static int tunnel_bts(struct rsvp_head *data) +{ + int n = data->tgenerator >> 5; + u32 b = 1 << (data->tgenerator & 0x1F); + + if (data->tmap[n] & b) + return 0; + data->tmap[n] |= b; + return 1; +} + +static void tunnel_recycle(struct rsvp_head *data) +{ + struct rsvp_session __rcu **sht = data->ht; + u32 tmap[256/32]; + int h1, h2; + + memset(tmap, 0, sizeof(tmap)); + + for (h1 = 0; h1 < 256; h1++) { + struct rsvp_session *s; + for (s = rtnl_dereference(sht[h1]); s; + s = rtnl_dereference(s->next)) { + for (h2 = 0; h2 <= 16; h2++) { + struct rsvp_filter *f; + + for (f = rtnl_dereference(s->ht[h2]); f; + f = rtnl_dereference(f->next)) { + if (f->tunnelhdr == 0) + continue; + data->tgenerator = f->res.classid; + tunnel_bts(data); + } + } + } + } + + memcpy(data->tmap, tmap, sizeof(tmap)); +} + +static u32 gen_tunnel(struct rsvp_head *data) +{ + int i, k; + + for (k = 0; k < 2; k++) { + for (i = 255; i > 0; i--) { + if (++data->tgenerator == 0) + data->tgenerator = 1; + if (tunnel_bts(data)) + return data->tgenerator; + } + tunnel_recycle(data); + } + return 0; +} + +static const struct nla_policy rsvp_policy[TCA_RSVP_MAX + 1] = { + [TCA_RSVP_CLASSID] = { .type = NLA_U32 }, + [TCA_RSVP_DST] = { .len = RSVP_DST_LEN * sizeof(u32) }, + [TCA_RSVP_SRC] = { .len = RSVP_DST_LEN * sizeof(u32) }, + [TCA_RSVP_PINFO] = { .len = sizeof(struct tc_rsvp_pinfo) }, +}; + +static int rsvp_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, + struct nlattr **tca, + void **arg, bool ovr, struct netlink_ext_ack *extack) +{ + struct rsvp_head *data = rtnl_dereference(tp->root); + struct rsvp_filter *f, *nfp; + struct rsvp_filter __rcu **fp; + struct rsvp_session *nsp, *s; + struct rsvp_session __rcu **sp; + struct tc_rsvp_pinfo *pinfo = NULL; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_RSVP_MAX + 1]; + struct tcf_exts e; + unsigned int h1, h2; + __be32 *dst; + int err; + + if (opt == NULL) + return handle ? -EINVAL : 0; + + err = nla_parse_nested(tb, TCA_RSVP_MAX, opt, rsvp_policy, NULL); + if (err < 0) + return err; + + err = tcf_exts_init(&e, TCA_RSVP_ACT, TCA_RSVP_POLICE); + if (err < 0) + return err; + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr, extack); + if (err < 0) + goto errout2; + + f = *arg; + if (f) { + /* Node exists: adjust only classid */ + struct rsvp_filter *n; + + if (f->handle != handle && handle) + goto errout2; + + n = kmemdup(f, sizeof(*f), GFP_KERNEL); + if (!n) { + err = -ENOMEM; + goto errout2; + } + + err = tcf_exts_init(&n->exts, TCA_RSVP_ACT, TCA_RSVP_POLICE); + if (err < 0) { + kfree(n); + goto errout2; + } + + if (tb[TCA_RSVP_CLASSID]) { + n->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]); + tcf_bind_filter(tp, &n->res, base); + } + + tcf_exts_change(&n->exts, &e); + rsvp_replace(tp, n, handle); + return 0; + } + + /* Now more serious part... */ + err = -EINVAL; + if (handle) + goto errout2; + if (tb[TCA_RSVP_DST] == NULL) + goto errout2; + + err = -ENOBUFS; + f = kzalloc(sizeof(struct rsvp_filter), GFP_KERNEL); + if (f == NULL) + goto errout2; + + err = tcf_exts_init(&f->exts, TCA_RSVP_ACT, TCA_RSVP_POLICE); + if (err < 0) + goto errout; + h2 = 16; + if (tb[TCA_RSVP_SRC]) { + memcpy(f->src, nla_data(tb[TCA_RSVP_SRC]), sizeof(f->src)); + h2 = hash_src(f->src); + } + if (tb[TCA_RSVP_PINFO]) { + pinfo = nla_data(tb[TCA_RSVP_PINFO]); + f->spi = pinfo->spi; + f->tunnelhdr = pinfo->tunnelhdr; + } + if (tb[TCA_RSVP_CLASSID]) + f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID]); + + dst = nla_data(tb[TCA_RSVP_DST]); + h1 = hash_dst(dst, pinfo ? pinfo->protocol : 0, pinfo ? pinfo->tunnelid : 0); + + err = -ENOMEM; + if ((f->handle = gen_handle(tp, h1 | (h2<<8))) == 0) + goto errout; + + if (f->tunnelhdr) { + err = -EINVAL; + if (f->res.classid > 255) + goto errout; + + err = -ENOMEM; + if (f->res.classid == 0 && + (f->res.classid = gen_tunnel(data)) == 0) + goto errout; + } + + for (sp = &data->ht[h1]; + (s = rtnl_dereference(*sp)) != NULL; + sp = &s->next) { + if (dst[RSVP_DST_LEN-1] == s->dst[RSVP_DST_LEN-1] && + pinfo && pinfo->protocol == s->protocol && + memcmp(&pinfo->dpi, &s->dpi, sizeof(s->dpi)) == 0 && +#if RSVP_DST_LEN == 4 + dst[0] == s->dst[0] && + dst[1] == s->dst[1] && + dst[2] == s->dst[2] && +#endif + pinfo->tunnelid == s->tunnelid) { + +insert: + /* OK, we found appropriate session */ + + fp = &s->ht[h2]; + + f->sess = s; + if (f->tunnelhdr == 0) + tcf_bind_filter(tp, &f->res, base); + + tcf_exts_change(&f->exts, &e); + + fp = &s->ht[h2]; + for (nfp = rtnl_dereference(*fp); nfp; + fp = &nfp->next, nfp = rtnl_dereference(*fp)) { + __u32 mask = nfp->spi.mask & f->spi.mask; + + if (mask != f->spi.mask) + break; + } + RCU_INIT_POINTER(f->next, nfp); + rcu_assign_pointer(*fp, f); + + *arg = f; + return 0; + } + } + + /* No session found. Create new one. */ + + err = -ENOBUFS; + s = kzalloc(sizeof(struct rsvp_session), GFP_KERNEL); + if (s == NULL) + goto errout; + memcpy(s->dst, dst, sizeof(s->dst)); + + if (pinfo) { + s->dpi = pinfo->dpi; + s->protocol = pinfo->protocol; + s->tunnelid = pinfo->tunnelid; + } + sp = &data->ht[h1]; + for (nsp = rtnl_dereference(*sp); nsp; + sp = &nsp->next, nsp = rtnl_dereference(*sp)) { + if ((nsp->dpi.mask & s->dpi.mask) != s->dpi.mask) + break; + } + RCU_INIT_POINTER(s->next, nsp); + rcu_assign_pointer(*sp, s); + + goto insert; + +errout: + tcf_exts_destroy(&f->exts); + kfree(f); +errout2: + tcf_exts_destroy(&e); + return err; +} + +static void rsvp_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct rsvp_head *head = rtnl_dereference(tp->root); + unsigned int h, h1; + + if (arg->stop) + return; + + for (h = 0; h < 256; h++) { + struct rsvp_session *s; + + for (s = rtnl_dereference(head->ht[h]); s; + s = rtnl_dereference(s->next)) { + for (h1 = 0; h1 <= 16; h1++) { + struct rsvp_filter *f; + + for (f = rtnl_dereference(s->ht[h1]); f; + f = rtnl_dereference(f->next)) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(tp, f, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } + } + } +} + +static int rsvp_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct rsvp_filter *f = fh; + struct rsvp_session *s; + struct nlattr *nest; + struct tc_rsvp_pinfo pinfo; + + if (f == NULL) + return skb->len; + s = f->sess; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (nla_put(skb, TCA_RSVP_DST, sizeof(s->dst), &s->dst)) + goto nla_put_failure; + pinfo.dpi = s->dpi; + pinfo.spi = f->spi; + pinfo.protocol = s->protocol; + pinfo.tunnelid = s->tunnelid; + pinfo.tunnelhdr = f->tunnelhdr; + pinfo.pad = 0; + if (nla_put(skb, TCA_RSVP_PINFO, sizeof(pinfo), &pinfo)) + goto nla_put_failure; + if (f->res.classid && + nla_put_u32(skb, TCA_RSVP_CLASSID, f->res.classid)) + goto nla_put_failure; + if (((f->handle >> 8) & 0xFF) != 16 && + nla_put(skb, TCA_RSVP_SRC, sizeof(f->src), f->src)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts) < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void rsvp_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct rsvp_filter *f = fh; + + if (f && f->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &f->res, base); + else + __tcf_unbind_filter(q, &f->res); + } +} + +static struct tcf_proto_ops RSVP_OPS __read_mostly = { + .kind = RSVP_ID, + .classify = rsvp_classify, + .init = rsvp_init, + .destroy = rsvp_destroy, + .get = rsvp_get, + .change = rsvp_change, + .delete = rsvp_delete, + .walk = rsvp_walk, + .dump = rsvp_dump, + .bind_class = rsvp_bind_class, + .owner = THIS_MODULE, +}; + +static int __init init_rsvp(void) +{ + return register_tcf_proto_ops(&RSVP_OPS); +} + +static void __exit exit_rsvp(void) +{ + unregister_tcf_proto_ops(&RSVP_OPS); +} + +module_init(init_rsvp) +module_exit(exit_rsvp) diff --git a/net/sched/cls_rsvp6.c b/net/sched/cls_rsvp6.c new file mode 100644 index 000000000..dd08aea2a --- /dev/null +++ b/net/sched/cls_rsvp6.c @@ -0,0 +1,28 @@ +/* + * net/sched/cls_rsvp6.c Special RSVP packet classifier for IPv6. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/ipv6.h> +#include <linux/skbuff.h> +#include <net/act_api.h> +#include <net/pkt_cls.h> +#include <net/netlink.h> + +#define RSVP_DST_LEN 4 +#define RSVP_ID "rsvp6" +#define RSVP_OPS cls_rsvp6_ops + +#include "cls_rsvp.h" +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c new file mode 100644 index 000000000..4070197f9 --- /dev/null +++ b/net/sched/cls_tcindex.c @@ -0,0 +1,698 @@ +/* + * net/sched/cls_tcindex.c Packet classifier for skb->tc_index + * + * Written 1998,1999 by Werner Almesberger, EPFL ICA + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <net/act_api.h> +#include <net/netlink.h> +#include <net/pkt_cls.h> +#include <net/sch_generic.h> + +/* + * Passing parameters to the root seems to be done more awkwardly than really + * necessary. At least, u32 doesn't seem to use such dirty hacks. To be + * verified. FIXME. + */ + +#define PERFECT_HASH_THRESHOLD 64 /* use perfect hash if not bigger */ +#define DEFAULT_HASH_SIZE 64 /* optimized for diffserv */ + + +struct tcindex_filter_result { + struct tcf_exts exts; + struct tcf_result res; + struct rcu_work rwork; +}; + +struct tcindex_filter { + u16 key; + struct tcindex_filter_result result; + struct tcindex_filter __rcu *next; + struct rcu_work rwork; +}; + + +struct tcindex_data { + struct tcindex_filter_result *perfect; /* perfect hash; NULL if none */ + struct tcindex_filter __rcu **h; /* imperfect hash; */ + struct tcf_proto *tp; + u16 mask; /* AND key with mask */ + u32 shift; /* shift ANDed key to the right */ + u32 hash; /* hash table size; 0 if undefined */ + u32 alloc_hash; /* allocated size */ + u32 fall_through; /* 0: only classify if explicit match */ + struct rcu_work rwork; +}; + +static inline int tcindex_filter_is_set(struct tcindex_filter_result *r) +{ + return tcf_exts_has_actions(&r->exts) || r->res.classid; +} + +static struct tcindex_filter_result *tcindex_lookup(struct tcindex_data *p, + u16 key) +{ + if (p->perfect) { + struct tcindex_filter_result *f = p->perfect + key; + + return tcindex_filter_is_set(f) ? f : NULL; + } else if (p->h) { + struct tcindex_filter __rcu **fp; + struct tcindex_filter *f; + + fp = &p->h[key % p->hash]; + for (f = rcu_dereference_bh_rtnl(*fp); + f; + fp = &f->next, f = rcu_dereference_bh_rtnl(*fp)) + if (f->key == key) + return &f->result; + } + + return NULL; +} + + +static int tcindex_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct tcindex_data *p = rcu_dereference_bh(tp->root); + struct tcindex_filter_result *f; + int key = (skb->tc_index & p->mask) >> p->shift; + + pr_debug("tcindex_classify(skb %p,tp %p,res %p),p %p\n", + skb, tp, res, p); + + f = tcindex_lookup(p, key); + if (!f) { + struct Qdisc *q = tcf_block_q(tp->chain->block); + + if (!p->fall_through) + return -1; + res->classid = TC_H_MAKE(TC_H_MAJ(q->handle), key); + res->class = 0; + pr_debug("alg 0x%x\n", res->classid); + return 0; + } + *res = f->res; + pr_debug("map 0x%x\n", res->classid); + + return tcf_exts_exec(skb, &f->exts, res); +} + + +static void *tcindex_get(struct tcf_proto *tp, u32 handle) +{ + struct tcindex_data *p = rtnl_dereference(tp->root); + struct tcindex_filter_result *r; + + pr_debug("tcindex_get(tp %p,handle 0x%08x)\n", tp, handle); + if (p->perfect && handle >= p->alloc_hash) + return NULL; + r = tcindex_lookup(p, handle); + return r && tcindex_filter_is_set(r) ? r : NULL; +} + +static int tcindex_init(struct tcf_proto *tp) +{ + struct tcindex_data *p; + + pr_debug("tcindex_init(tp %p)\n", tp); + p = kzalloc(sizeof(struct tcindex_data), GFP_KERNEL); + if (!p) + return -ENOMEM; + + p->mask = 0xffff; + p->hash = DEFAULT_HASH_SIZE; + p->fall_through = 1; + + rcu_assign_pointer(tp->root, p); + return 0; +} + +static void __tcindex_destroy_rexts(struct tcindex_filter_result *r) +{ + tcf_exts_destroy(&r->exts); + tcf_exts_put_net(&r->exts); +} + +static void tcindex_destroy_rexts_work(struct work_struct *work) +{ + struct tcindex_filter_result *r; + + r = container_of(to_rcu_work(work), + struct tcindex_filter_result, + rwork); + rtnl_lock(); + __tcindex_destroy_rexts(r); + rtnl_unlock(); +} + +static void __tcindex_destroy_fexts(struct tcindex_filter *f) +{ + tcf_exts_destroy(&f->result.exts); + tcf_exts_put_net(&f->result.exts); + kfree(f); +} + +static void tcindex_destroy_fexts_work(struct work_struct *work) +{ + struct tcindex_filter *f = container_of(to_rcu_work(work), + struct tcindex_filter, + rwork); + + rtnl_lock(); + __tcindex_destroy_fexts(f); + rtnl_unlock(); +} + +static int tcindex_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct tcindex_data *p = rtnl_dereference(tp->root); + struct tcindex_filter_result *r = arg; + struct tcindex_filter __rcu **walk; + struct tcindex_filter *f = NULL; + + pr_debug("tcindex_delete(tp %p,arg %p),p %p\n", tp, arg, p); + if (p->perfect) { + if (!r->res.class) + return -ENOENT; + } else { + int i; + + for (i = 0; i < p->hash; i++) { + walk = p->h + i; + for (f = rtnl_dereference(*walk); f; + walk = &f->next, f = rtnl_dereference(*walk)) { + if (&f->result == r) + goto found; + } + } + return -ENOENT; + +found: + rcu_assign_pointer(*walk, rtnl_dereference(f->next)); + } + tcf_unbind_filter(tp, &r->res); + /* all classifiers are required to call tcf_exts_destroy() after rcu + * grace period, since converted-to-rcu actions are relying on that + * in cleanup() callback + */ + if (f) { + if (tcf_exts_get_net(&f->result.exts)) + tcf_queue_work(&f->rwork, tcindex_destroy_fexts_work); + else + __tcindex_destroy_fexts(f); + } else { + if (tcf_exts_get_net(&r->exts)) + tcf_queue_work(&r->rwork, tcindex_destroy_rexts_work); + else + __tcindex_destroy_rexts(r); + } + + *last = false; + return 0; +} + +static void tcindex_destroy_work(struct work_struct *work) +{ + struct tcindex_data *p = container_of(to_rcu_work(work), + struct tcindex_data, + rwork); + + kfree(p->perfect); + kfree(p->h); + kfree(p); +} + +static inline int +valid_perfect_hash(struct tcindex_data *p) +{ + return p->hash > (p->mask >> p->shift); +} + +static const struct nla_policy tcindex_policy[TCA_TCINDEX_MAX + 1] = { + [TCA_TCINDEX_HASH] = { .type = NLA_U32 }, + [TCA_TCINDEX_MASK] = { .type = NLA_U16 }, + [TCA_TCINDEX_SHIFT] = { .type = NLA_U32 }, + [TCA_TCINDEX_FALL_THROUGH] = { .type = NLA_U32 }, + [TCA_TCINDEX_CLASSID] = { .type = NLA_U32 }, +}; + +static int tcindex_filter_result_init(struct tcindex_filter_result *r) +{ + memset(r, 0, sizeof(*r)); + return tcf_exts_init(&r->exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); +} + +static void tcindex_partial_destroy_work(struct work_struct *work) +{ + struct tcindex_data *p = container_of(to_rcu_work(work), + struct tcindex_data, + rwork); + + kfree(p->perfect); + kfree(p); +} + +static void tcindex_free_perfect_hash(struct tcindex_data *cp) +{ + int i; + + for (i = 0; i < cp->hash; i++) + tcf_exts_destroy(&cp->perfect[i].exts); + kfree(cp->perfect); +} + +static int tcindex_alloc_perfect_hash(struct net *net, struct tcindex_data *cp) +{ + int i, err = 0; + + cp->perfect = kcalloc(cp->hash, sizeof(struct tcindex_filter_result), + GFP_KERNEL | __GFP_NOWARN); + if (!cp->perfect) + return -ENOMEM; + + for (i = 0; i < cp->hash; i++) { + err = tcf_exts_init(&cp->perfect[i].exts, + TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); + if (err < 0) + goto errout; +#ifdef CONFIG_NET_CLS_ACT + cp->perfect[i].exts.net = net; +#endif + } + + return 0; + +errout: + tcindex_free_perfect_hash(cp); + return err; +} + +static int +tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, + u32 handle, struct tcindex_data *p, + struct tcindex_filter_result *r, struct nlattr **tb, + struct nlattr *est, bool ovr, struct netlink_ext_ack *extack) +{ + struct tcindex_filter_result new_filter_result, *old_r = r; + struct tcindex_data *cp = NULL, *oldp; + struct tcindex_filter *f = NULL; /* make gcc behave */ + struct tcf_result cr = {}; + int err, balloc = 0; + struct tcf_exts e; + + err = tcf_exts_init(&e, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); + if (err < 0) + return err; + err = tcf_exts_validate(net, tp, tb, est, &e, ovr, extack); + if (err < 0) + goto errout; + + err = -ENOMEM; + /* tcindex_data attributes must look atomic to classifier/lookup so + * allocate new tcindex data and RCU assign it onto root. Keeping + * perfect hash and hash pointers from old data. + */ + cp = kzalloc(sizeof(*cp), GFP_KERNEL); + if (!cp) + goto errout; + + cp->mask = p->mask; + cp->shift = p->shift; + cp->hash = p->hash; + cp->alloc_hash = p->alloc_hash; + cp->fall_through = p->fall_through; + cp->tp = tp; + + if (tb[TCA_TCINDEX_HASH]) + cp->hash = nla_get_u32(tb[TCA_TCINDEX_HASH]); + + if (tb[TCA_TCINDEX_MASK]) + cp->mask = nla_get_u16(tb[TCA_TCINDEX_MASK]); + + if (tb[TCA_TCINDEX_SHIFT]) { + cp->shift = nla_get_u32(tb[TCA_TCINDEX_SHIFT]); + if (cp->shift > 16) { + err = -EINVAL; + goto errout; + } + } + if (!cp->hash) { + /* Hash not specified, use perfect hash if the upper limit + * of the hashing index is below the threshold. + */ + if ((cp->mask >> cp->shift) < PERFECT_HASH_THRESHOLD) + cp->hash = (cp->mask >> cp->shift) + 1; + else + cp->hash = DEFAULT_HASH_SIZE; + } + + if (p->perfect) { + int i; + + if (tcindex_alloc_perfect_hash(net, cp) < 0) + goto errout; + cp->alloc_hash = cp->hash; + for (i = 0; i < min(cp->hash, p->hash); i++) + cp->perfect[i].res = p->perfect[i].res; + balloc = 1; + } + cp->h = p->h; + + err = tcindex_filter_result_init(&new_filter_result); + if (err < 0) + goto errout_alloc; + if (old_r) + cr = r->res; + + err = -EBUSY; + + /* Hash already allocated, make sure that we still meet the + * requirements for the allocated hash. + */ + if (cp->perfect) { + if (!valid_perfect_hash(cp) || + cp->hash > cp->alloc_hash) + goto errout_alloc; + } else if (cp->h && cp->hash != cp->alloc_hash) { + goto errout_alloc; + } + + err = -EINVAL; + if (tb[TCA_TCINDEX_FALL_THROUGH]) + cp->fall_through = nla_get_u32(tb[TCA_TCINDEX_FALL_THROUGH]); + + if (!cp->perfect && !cp->h) + cp->alloc_hash = cp->hash; + + /* Note: this could be as restrictive as if (handle & ~(mask >> shift)) + * but then, we'd fail handles that may become valid after some future + * mask change. While this is extremely unlikely to ever matter, + * the check below is safer (and also more backwards-compatible). + */ + if (cp->perfect || valid_perfect_hash(cp)) + if (handle >= cp->alloc_hash) + goto errout_alloc; + + + err = -ENOMEM; + if (!cp->perfect && !cp->h) { + if (valid_perfect_hash(cp)) { + if (tcindex_alloc_perfect_hash(net, cp) < 0) + goto errout_alloc; + balloc = 1; + } else { + struct tcindex_filter __rcu **hash; + + hash = kcalloc(cp->hash, + sizeof(struct tcindex_filter *), + GFP_KERNEL); + + if (!hash) + goto errout_alloc; + + cp->h = hash; + balloc = 2; + } + } + + if (cp->perfect) + r = cp->perfect + handle; + else + r = tcindex_lookup(cp, handle) ? : &new_filter_result; + + if (r == &new_filter_result) { + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + goto errout_alloc; + f->key = handle; + f->next = NULL; + err = tcindex_filter_result_init(&f->result); + if (err < 0) { + kfree(f); + goto errout_alloc; + } + } + + if (tb[TCA_TCINDEX_CLASSID]) { + cr.classid = nla_get_u32(tb[TCA_TCINDEX_CLASSID]); + tcf_bind_filter(tp, &cr, base); + } + + if (old_r && old_r != r) { + err = tcindex_filter_result_init(old_r); + if (err < 0) { + kfree(f); + goto errout_alloc; + } + } + + oldp = p; + r->res = cr; + tcf_exts_change(&r->exts, &e); + + rcu_assign_pointer(tp->root, cp); + + if (r == &new_filter_result) { + struct tcindex_filter *nfp; + struct tcindex_filter __rcu **fp; + + f->result.res = r->res; + tcf_exts_change(&f->result.exts, &r->exts); + + fp = cp->h + (handle % cp->hash); + for (nfp = rtnl_dereference(*fp); + nfp; + fp = &nfp->next, nfp = rtnl_dereference(*fp)) + ; /* nothing */ + + rcu_assign_pointer(*fp, f); + } else { + tcf_exts_destroy(&new_filter_result.exts); + } + + if (oldp) + tcf_queue_work(&oldp->rwork, tcindex_partial_destroy_work); + return 0; + +errout_alloc: + if (balloc == 1) + tcindex_free_perfect_hash(cp); + else if (balloc == 2) + kfree(cp->h); + tcf_exts_destroy(&new_filter_result.exts); +errout: + kfree(cp); + tcf_exts_destroy(&e); + return err; +} + +static int +tcindex_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, u32 handle, + struct nlattr **tca, void **arg, bool ovr, + struct netlink_ext_ack *extack) +{ + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_TCINDEX_MAX + 1]; + struct tcindex_data *p = rtnl_dereference(tp->root); + struct tcindex_filter_result *r = *arg; + int err; + + pr_debug("tcindex_change(tp %p,handle 0x%08x,tca %p,arg %p),opt %p," + "p %p,r %p,*arg %p\n", + tp, handle, tca, arg, opt, p, r, arg ? *arg : NULL); + + if (!opt) + return 0; + + err = nla_parse_nested(tb, TCA_TCINDEX_MAX, opt, tcindex_policy, NULL); + if (err < 0) + return err; + + return tcindex_set_parms(net, tp, base, handle, p, r, tb, + tca[TCA_RATE], ovr, extack); +} + +static void tcindex_walk(struct tcf_proto *tp, struct tcf_walker *walker) +{ + struct tcindex_data *p = rtnl_dereference(tp->root); + struct tcindex_filter *f, *next; + int i; + + pr_debug("tcindex_walk(tp %p,walker %p),p %p\n", tp, walker, p); + if (p->perfect) { + for (i = 0; i < p->hash; i++) { + if (!p->perfect[i].res.class) + continue; + if (walker->count >= walker->skip) { + if (walker->fn(tp, p->perfect + i, walker) < 0) { + walker->stop = 1; + return; + } + } + walker->count++; + } + } + if (!p->h) + return; + for (i = 0; i < p->hash; i++) { + for (f = rtnl_dereference(p->h[i]); f; f = next) { + next = rtnl_dereference(f->next); + if (walker->count >= walker->skip) { + if (walker->fn(tp, &f->result, walker) < 0) { + walker->stop = 1; + return; + } + } + walker->count++; + } + } +} + +static void tcindex_destroy(struct tcf_proto *tp, + struct netlink_ext_ack *extack) +{ + struct tcindex_data *p = rtnl_dereference(tp->root); + int i; + + pr_debug("tcindex_destroy(tp %p),p %p\n", tp, p); + + if (p->perfect) { + for (i = 0; i < p->hash; i++) { + struct tcindex_filter_result *r = p->perfect + i; + + tcf_unbind_filter(tp, &r->res); + if (tcf_exts_get_net(&r->exts)) + tcf_queue_work(&r->rwork, + tcindex_destroy_rexts_work); + else + __tcindex_destroy_rexts(r); + } + } + + for (i = 0; p->h && i < p->hash; i++) { + struct tcindex_filter *f, *next; + bool last; + + for (f = rtnl_dereference(p->h[i]); f; f = next) { + next = rtnl_dereference(f->next); + tcindex_delete(tp, &f->result, &last, NULL); + } + } + + tcf_queue_work(&p->rwork, tcindex_destroy_work); +} + + +static int tcindex_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct tcindex_data *p = rtnl_dereference(tp->root); + struct tcindex_filter_result *r = fh; + struct nlattr *nest; + + pr_debug("tcindex_dump(tp %p,fh %p,skb %p,t %p),p %p,r %p\n", + tp, fh, skb, t, p, r); + pr_debug("p->perfect %p p->h %p\n", p->perfect, p->h); + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (!fh) { + t->tcm_handle = ~0; /* whatever ... */ + if (nla_put_u32(skb, TCA_TCINDEX_HASH, p->hash) || + nla_put_u16(skb, TCA_TCINDEX_MASK, p->mask) || + nla_put_u32(skb, TCA_TCINDEX_SHIFT, p->shift) || + nla_put_u32(skb, TCA_TCINDEX_FALL_THROUGH, p->fall_through)) + goto nla_put_failure; + nla_nest_end(skb, nest); + } else { + if (p->perfect) { + t->tcm_handle = r - p->perfect; + } else { + struct tcindex_filter *f; + struct tcindex_filter __rcu **fp; + int i; + + t->tcm_handle = 0; + for (i = 0; !t->tcm_handle && i < p->hash; i++) { + fp = &p->h[i]; + for (f = rtnl_dereference(*fp); + !t->tcm_handle && f; + fp = &f->next, f = rtnl_dereference(*fp)) { + if (&f->result == r) + t->tcm_handle = f->key; + } + } + } + pr_debug("handle = %d\n", t->tcm_handle); + if (r->res.class && + nla_put_u32(skb, TCA_TCINDEX_CLASSID, r->res.classid)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &r->exts) < 0) + goto nla_put_failure; + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &r->exts) < 0) + goto nla_put_failure; + } + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static void tcindex_bind_class(void *fh, u32 classid, unsigned long cl, + void *q, unsigned long base) +{ + struct tcindex_filter_result *r = fh; + + if (r && r->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &r->res, base); + else + __tcf_unbind_filter(q, &r->res); + } +} + +static struct tcf_proto_ops cls_tcindex_ops __read_mostly = { + .kind = "tcindex", + .classify = tcindex_classify, + .init = tcindex_init, + .destroy = tcindex_destroy, + .get = tcindex_get, + .change = tcindex_change, + .delete = tcindex_delete, + .walk = tcindex_walk, + .dump = tcindex_dump, + .bind_class = tcindex_bind_class, + .owner = THIS_MODULE, +}; + +static int __init init_tcindex(void) +{ + return register_tcf_proto_ops(&cls_tcindex_ops); +} + +static void __exit exit_tcindex(void) +{ + unregister_tcf_proto_ops(&cls_tcindex_ops); +} + +module_init(init_tcindex) +module_exit(exit_tcindex) +MODULE_LICENSE("GPL"); diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c new file mode 100644 index 000000000..d30256ac3 --- /dev/null +++ b/net/sched/cls_u32.c @@ -0,0 +1,1507 @@ +/* + * net/sched/cls_u32.c Ugly (or Universal) 32bit key Packet Classifier. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + * The filters are packed to hash tables of key nodes + * with a set of 32bit key/mask pairs at every node. + * Nodes reference next level hash tables etc. + * + * This scheme is the best universal classifier I managed to + * invent; it is not super-fast, but it is not slow (provided you + * program it correctly), and general enough. And its relative + * speed grows as the number of rules becomes larger. + * + * It seems that it represents the best middle point between + * speed and manageability both by human and by machine. + * + * It is especially useful for link sharing combined with QoS; + * pure RSVP doesn't need such a general approach and can use + * much simpler (and faster) schemes, sort of cls_rsvp.c. + * + * JHS: We should remove the CONFIG_NET_CLS_IND from here + * eventually when the meta match extension is made available + * + * nfmark match added by Catalin(ux aka Dino) BOIE <catab at umbrella.ro> + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/percpu.h> +#include <linux/rtnetlink.h> +#include <linux/skbuff.h> +#include <linux/bitmap.h> +#include <linux/netdevice.h> +#include <linux/hash.h> +#include <net/netlink.h> +#include <net/act_api.h> +#include <net/pkt_cls.h> +#include <linux/idr.h> + +struct tc_u_knode { + struct tc_u_knode __rcu *next; + u32 handle; + struct tc_u_hnode __rcu *ht_up; + struct tcf_exts exts; +#ifdef CONFIG_NET_CLS_IND + int ifindex; +#endif + u8 fshift; + struct tcf_result res; + struct tc_u_hnode __rcu *ht_down; +#ifdef CONFIG_CLS_U32_PERF + struct tc_u32_pcnt __percpu *pf; +#endif + u32 flags; + unsigned int in_hw_count; +#ifdef CONFIG_CLS_U32_MARK + u32 val; + u32 mask; + u32 __percpu *pcpu_success; +#endif + struct tcf_proto *tp; + struct rcu_work rwork; + /* The 'sel' field MUST be the last field in structure to allow for + * tc_u32_keys allocated at end of structure. + */ + struct tc_u32_sel sel; +}; + +struct tc_u_hnode { + struct tc_u_hnode __rcu *next; + u32 handle; + u32 prio; + struct tc_u_common *tp_c; + int refcnt; + unsigned int divisor; + struct idr handle_idr; + struct rcu_head rcu; + u32 flags; + /* The 'ht' field MUST be the last field in structure to allow for + * more entries allocated at end of structure. + */ + struct tc_u_knode __rcu *ht[1]; +}; + +struct tc_u_common { + struct tc_u_hnode __rcu *hlist; + void *ptr; + int refcnt; + struct idr handle_idr; + struct hlist_node hnode; + struct rcu_head rcu; +}; + +static inline unsigned int u32_hash_fold(__be32 key, + const struct tc_u32_sel *sel, + u8 fshift) +{ + unsigned int h = ntohl(key & sel->hmask) >> fshift; + + return h; +} + +static int u32_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct { + struct tc_u_knode *knode; + unsigned int off; + } stack[TC_U32_MAXDEPTH]; + + struct tc_u_hnode *ht = rcu_dereference_bh(tp->root); + unsigned int off = skb_network_offset(skb); + struct tc_u_knode *n; + int sdepth = 0; + int off2 = 0; + int sel = 0; +#ifdef CONFIG_CLS_U32_PERF + int j; +#endif + int i, r; + +next_ht: + n = rcu_dereference_bh(ht->ht[sel]); + +next_knode: + if (n) { + struct tc_u32_key *key = n->sel.keys; + +#ifdef CONFIG_CLS_U32_PERF + __this_cpu_inc(n->pf->rcnt); + j = 0; +#endif + + if (tc_skip_sw(n->flags)) { + n = rcu_dereference_bh(n->next); + goto next_knode; + } + +#ifdef CONFIG_CLS_U32_MARK + if ((skb->mark & n->mask) != n->val) { + n = rcu_dereference_bh(n->next); + goto next_knode; + } else { + __this_cpu_inc(*n->pcpu_success); + } +#endif + + for (i = n->sel.nkeys; i > 0; i--, key++) { + int toff = off + key->off + (off2 & key->offmask); + __be32 *data, hdata; + + if (skb_headroom(skb) + toff > INT_MAX) + goto out; + + data = skb_header_pointer(skb, toff, 4, &hdata); + if (!data) + goto out; + if ((*data ^ key->val) & key->mask) { + n = rcu_dereference_bh(n->next); + goto next_knode; + } +#ifdef CONFIG_CLS_U32_PERF + __this_cpu_inc(n->pf->kcnts[j]); + j++; +#endif + } + + ht = rcu_dereference_bh(n->ht_down); + if (!ht) { +check_terminal: + if (n->sel.flags & TC_U32_TERMINAL) { + + *res = n->res; +#ifdef CONFIG_NET_CLS_IND + if (!tcf_match_indev(skb, n->ifindex)) { + n = rcu_dereference_bh(n->next); + goto next_knode; + } +#endif +#ifdef CONFIG_CLS_U32_PERF + __this_cpu_inc(n->pf->rhit); +#endif + r = tcf_exts_exec(skb, &n->exts, res); + if (r < 0) { + n = rcu_dereference_bh(n->next); + goto next_knode; + } + + return r; + } + n = rcu_dereference_bh(n->next); + goto next_knode; + } + + /* PUSH */ + if (sdepth >= TC_U32_MAXDEPTH) + goto deadloop; + stack[sdepth].knode = n; + stack[sdepth].off = off; + sdepth++; + + ht = rcu_dereference_bh(n->ht_down); + sel = 0; + if (ht->divisor) { + __be32 *data, hdata; + + data = skb_header_pointer(skb, off + n->sel.hoff, 4, + &hdata); + if (!data) + goto out; + sel = ht->divisor & u32_hash_fold(*data, &n->sel, + n->fshift); + } + if (!(n->sel.flags & (TC_U32_VAROFFSET | TC_U32_OFFSET | TC_U32_EAT))) + goto next_ht; + + if (n->sel.flags & (TC_U32_OFFSET | TC_U32_VAROFFSET)) { + off2 = n->sel.off + 3; + if (n->sel.flags & TC_U32_VAROFFSET) { + __be16 *data, hdata; + + data = skb_header_pointer(skb, + off + n->sel.offoff, + 2, &hdata); + if (!data) + goto out; + off2 += ntohs(n->sel.offmask & *data) >> + n->sel.offshift; + } + off2 &= ~3; + } + if (n->sel.flags & TC_U32_EAT) { + off += off2; + off2 = 0; + } + + if (off < skb->len) + goto next_ht; + } + + /* POP */ + if (sdepth--) { + n = stack[sdepth].knode; + ht = rcu_dereference_bh(n->ht_up); + off = stack[sdepth].off; + goto check_terminal; + } +out: + return -1; + +deadloop: + net_warn_ratelimited("cls_u32: dead loop\n"); + return -1; +} + +static struct tc_u_hnode *u32_lookup_ht(struct tc_u_common *tp_c, u32 handle) +{ + struct tc_u_hnode *ht; + + for (ht = rtnl_dereference(tp_c->hlist); + ht; + ht = rtnl_dereference(ht->next)) + if (ht->handle == handle) + break; + + return ht; +} + +static struct tc_u_knode *u32_lookup_key(struct tc_u_hnode *ht, u32 handle) +{ + unsigned int sel; + struct tc_u_knode *n = NULL; + + sel = TC_U32_HASH(handle); + if (sel > ht->divisor) + goto out; + + for (n = rtnl_dereference(ht->ht[sel]); + n; + n = rtnl_dereference(n->next)) + if (n->handle == handle) + break; +out: + return n; +} + + +static void *u32_get(struct tcf_proto *tp, u32 handle) +{ + struct tc_u_hnode *ht; + struct tc_u_common *tp_c = tp->data; + + if (TC_U32_HTID(handle) == TC_U32_ROOT) + ht = rtnl_dereference(tp->root); + else + ht = u32_lookup_ht(tp_c, TC_U32_HTID(handle)); + + if (!ht) + return NULL; + + if (TC_U32_KEY(handle) == 0) + return ht; + + return u32_lookup_key(ht, handle); +} + +/* Protected by rtnl lock */ +static u32 gen_new_htid(struct tc_u_common *tp_c, struct tc_u_hnode *ptr) +{ + int id = idr_alloc_cyclic(&tp_c->handle_idr, ptr, 1, 0x7FF, GFP_KERNEL); + if (id < 0) + return 0; + return (id | 0x800U) << 20; +} + +static struct hlist_head *tc_u_common_hash; + +#define U32_HASH_SHIFT 10 +#define U32_HASH_SIZE (1 << U32_HASH_SHIFT) + +static void *tc_u_common_ptr(const struct tcf_proto *tp) +{ + struct tcf_block *block = tp->chain->block; + + /* The block sharing is currently supported only + * for classless qdiscs. In that case we use block + * for tc_u_common identification. In case the + * block is not shared, block->q is a valid pointer + * and we can use that. That works for classful qdiscs. + */ + if (tcf_block_shared(block)) + return block; + else + return block->q; +} + +static unsigned int tc_u_hash(const struct tcf_proto *tp) +{ + return hash_ptr(tc_u_common_ptr(tp), U32_HASH_SHIFT); +} + +static struct tc_u_common *tc_u_common_find(const struct tcf_proto *tp) +{ + struct tc_u_common *tc; + unsigned int h; + + h = tc_u_hash(tp); + hlist_for_each_entry(tc, &tc_u_common_hash[h], hnode) { + if (tc->ptr == tc_u_common_ptr(tp)) + return tc; + } + return NULL; +} + +static int u32_init(struct tcf_proto *tp) +{ + struct tc_u_hnode *root_ht; + struct tc_u_common *tp_c; + unsigned int h; + + tp_c = tc_u_common_find(tp); + + root_ht = kzalloc(sizeof(*root_ht), GFP_KERNEL); + if (root_ht == NULL) + return -ENOBUFS; + + root_ht->refcnt++; + root_ht->handle = tp_c ? gen_new_htid(tp_c, root_ht) : 0x80000000; + root_ht->prio = tp->prio; + idr_init(&root_ht->handle_idr); + + if (tp_c == NULL) { + tp_c = kzalloc(sizeof(*tp_c), GFP_KERNEL); + if (tp_c == NULL) { + kfree(root_ht); + return -ENOBUFS; + } + tp_c->ptr = tc_u_common_ptr(tp); + INIT_HLIST_NODE(&tp_c->hnode); + idr_init(&tp_c->handle_idr); + + h = tc_u_hash(tp); + hlist_add_head(&tp_c->hnode, &tc_u_common_hash[h]); + } + + tp_c->refcnt++; + RCU_INIT_POINTER(root_ht->next, tp_c->hlist); + rcu_assign_pointer(tp_c->hlist, root_ht); + root_ht->tp_c = tp_c; + + root_ht->refcnt++; + rcu_assign_pointer(tp->root, root_ht); + tp->data = tp_c; + return 0; +} + +static void __u32_destroy_key(struct tc_u_knode *n) +{ + struct tc_u_hnode *ht = rtnl_dereference(n->ht_down); + + tcf_exts_destroy(&n->exts); + if (ht && --ht->refcnt == 0) + kfree(ht); + kfree(n); +} + +static void u32_destroy_key(struct tcf_proto *tp, struct tc_u_knode *n, + bool free_pf) +{ + tcf_exts_put_net(&n->exts); +#ifdef CONFIG_CLS_U32_PERF + if (free_pf) + free_percpu(n->pf); +#endif +#ifdef CONFIG_CLS_U32_MARK + if (free_pf) + free_percpu(n->pcpu_success); +#endif + __u32_destroy_key(n); +} + +/* u32_delete_key_rcu should be called when free'ing a copied + * version of a tc_u_knode obtained from u32_init_knode(). When + * copies are obtained from u32_init_knode() the statistics are + * shared between the old and new copies to allow readers to + * continue to update the statistics during the copy. To support + * this the u32_delete_key_rcu variant does not free the percpu + * statistics. + */ +static void u32_delete_key_work(struct work_struct *work) +{ + struct tc_u_knode *key = container_of(to_rcu_work(work), + struct tc_u_knode, + rwork); + rtnl_lock(); + u32_destroy_key(key->tp, key, false); + rtnl_unlock(); +} + +/* u32_delete_key_freepf_rcu is the rcu callback variant + * that free's the entire structure including the statistics + * percpu variables. Only use this if the key is not a copy + * returned by u32_init_knode(). See u32_delete_key_rcu() + * for the variant that should be used with keys return from + * u32_init_knode() + */ +static void u32_delete_key_freepf_work(struct work_struct *work) +{ + struct tc_u_knode *key = container_of(to_rcu_work(work), + struct tc_u_knode, + rwork); + rtnl_lock(); + u32_destroy_key(key->tp, key, true); + rtnl_unlock(); +} + +static int u32_delete_key(struct tcf_proto *tp, struct tc_u_knode *key) +{ + struct tc_u_knode __rcu **kp; + struct tc_u_knode *pkp; + struct tc_u_hnode *ht = rtnl_dereference(key->ht_up); + + if (ht) { + kp = &ht->ht[TC_U32_HASH(key->handle)]; + for (pkp = rtnl_dereference(*kp); pkp; + kp = &pkp->next, pkp = rtnl_dereference(*kp)) { + if (pkp == key) { + RCU_INIT_POINTER(*kp, key->next); + + tcf_unbind_filter(tp, &key->res); + idr_remove(&ht->handle_idr, key->handle); + tcf_exts_get_net(&key->exts); + tcf_queue_work(&key->rwork, u32_delete_key_freepf_work); + return 0; + } + } + } + WARN_ON(1); + return 0; +} + +static void u32_clear_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, + struct netlink_ext_ack *extack) +{ + struct tcf_block *block = tp->chain->block; + struct tc_cls_u32_offload cls_u32 = {}; + + tc_cls_common_offload_init(&cls_u32.common, tp, h->flags, extack); + cls_u32.command = TC_CLSU32_DELETE_HNODE; + cls_u32.hnode.divisor = h->divisor; + cls_u32.hnode.handle = h->handle; + cls_u32.hnode.prio = h->prio; + + tc_setup_cb_call(block, NULL, TC_SETUP_CLSU32, &cls_u32, false); +} + +static int u32_replace_hw_hnode(struct tcf_proto *tp, struct tc_u_hnode *h, + u32 flags, struct netlink_ext_ack *extack) +{ + struct tcf_block *block = tp->chain->block; + struct tc_cls_u32_offload cls_u32 = {}; + bool skip_sw = tc_skip_sw(flags); + bool offloaded = false; + int err; + + tc_cls_common_offload_init(&cls_u32.common, tp, flags, extack); + cls_u32.command = TC_CLSU32_NEW_HNODE; + cls_u32.hnode.divisor = h->divisor; + cls_u32.hnode.handle = h->handle; + cls_u32.hnode.prio = h->prio; + + err = tc_setup_cb_call(block, NULL, TC_SETUP_CLSU32, &cls_u32, skip_sw); + if (err < 0) { + u32_clear_hw_hnode(tp, h, NULL); + return err; + } else if (err > 0) { + offloaded = true; + } + + if (skip_sw && !offloaded) + return -EINVAL; + + return 0; +} + +static void u32_remove_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n, + struct netlink_ext_ack *extack) +{ + struct tcf_block *block = tp->chain->block; + struct tc_cls_u32_offload cls_u32 = {}; + + tc_cls_common_offload_init(&cls_u32.common, tp, n->flags, extack); + cls_u32.command = TC_CLSU32_DELETE_KNODE; + cls_u32.knode.handle = n->handle; + + tc_setup_cb_call(block, NULL, TC_SETUP_CLSU32, &cls_u32, false); + tcf_block_offload_dec(block, &n->flags); +} + +static int u32_replace_hw_knode(struct tcf_proto *tp, struct tc_u_knode *n, + u32 flags, struct netlink_ext_ack *extack) +{ + struct tc_u_hnode *ht = rtnl_dereference(n->ht_down); + struct tcf_block *block = tp->chain->block; + struct tc_cls_u32_offload cls_u32 = {}; + bool skip_sw = tc_skip_sw(flags); + int err; + + tc_cls_common_offload_init(&cls_u32.common, tp, flags, extack); + cls_u32.command = TC_CLSU32_REPLACE_KNODE; + cls_u32.knode.handle = n->handle; + cls_u32.knode.fshift = n->fshift; +#ifdef CONFIG_CLS_U32_MARK + cls_u32.knode.val = n->val; + cls_u32.knode.mask = n->mask; +#else + cls_u32.knode.val = 0; + cls_u32.knode.mask = 0; +#endif + cls_u32.knode.sel = &n->sel; + cls_u32.knode.exts = &n->exts; + if (n->ht_down) + cls_u32.knode.link_handle = ht->handle; + + err = tc_setup_cb_call(block, NULL, TC_SETUP_CLSU32, &cls_u32, skip_sw); + if (err < 0) { + u32_remove_hw_knode(tp, n, NULL); + return err; + } else if (err > 0) { + n->in_hw_count = err; + tcf_block_offload_inc(block, &n->flags); + } + + if (skip_sw && !(n->flags & TCA_CLS_FLAGS_IN_HW)) + return -EINVAL; + + return 0; +} + +static void u32_clear_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht, + struct netlink_ext_ack *extack) +{ + struct tc_u_knode *n; + unsigned int h; + + for (h = 0; h <= ht->divisor; h++) { + while ((n = rtnl_dereference(ht->ht[h])) != NULL) { + RCU_INIT_POINTER(ht->ht[h], + rtnl_dereference(n->next)); + tcf_unbind_filter(tp, &n->res); + u32_remove_hw_knode(tp, n, extack); + idr_remove(&ht->handle_idr, n->handle); + if (tcf_exts_get_net(&n->exts)) + tcf_queue_work(&n->rwork, u32_delete_key_freepf_work); + else + u32_destroy_key(n->tp, n, true); + } + } +} + +static int u32_destroy_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht, + struct netlink_ext_ack *extack) +{ + struct tc_u_common *tp_c = tp->data; + struct tc_u_hnode __rcu **hn; + struct tc_u_hnode *phn; + + WARN_ON(--ht->refcnt); + + u32_clear_hnode(tp, ht, extack); + + hn = &tp_c->hlist; + for (phn = rtnl_dereference(*hn); + phn; + hn = &phn->next, phn = rtnl_dereference(*hn)) { + if (phn == ht) { + u32_clear_hw_hnode(tp, ht, extack); + idr_destroy(&ht->handle_idr); + idr_remove(&tp_c->handle_idr, ht->handle); + RCU_INIT_POINTER(*hn, ht->next); + kfree_rcu(ht, rcu); + return 0; + } + } + + return -ENOENT; +} + +static bool ht_empty(struct tc_u_hnode *ht) +{ + unsigned int h; + + for (h = 0; h <= ht->divisor; h++) + if (rcu_access_pointer(ht->ht[h])) + return false; + + return true; +} + +static void u32_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack) +{ + struct tc_u_common *tp_c = tp->data; + struct tc_u_hnode *root_ht = rtnl_dereference(tp->root); + + WARN_ON(root_ht == NULL); + + if (root_ht && --root_ht->refcnt == 1) + u32_destroy_hnode(tp, root_ht, extack); + + if (--tp_c->refcnt == 0) { + struct tc_u_hnode *ht; + + hlist_del(&tp_c->hnode); + + while ((ht = rtnl_dereference(tp_c->hlist)) != NULL) { + u32_clear_hnode(tp, ht, extack); + RCU_INIT_POINTER(tp_c->hlist, ht->next); + + /* u32_destroy_key() will later free ht for us, if it's + * still referenced by some knode + */ + if (--ht->refcnt == 0) + kfree_rcu(ht, rcu); + } + + idr_destroy(&tp_c->handle_idr); + kfree(tp_c); + } + + tp->data = NULL; +} + +static int u32_delete(struct tcf_proto *tp, void *arg, bool *last, + struct netlink_ext_ack *extack) +{ + struct tc_u_hnode *ht = arg; + struct tc_u_hnode *root_ht = rtnl_dereference(tp->root); + struct tc_u_common *tp_c = tp->data; + int ret = 0; + + if (ht == NULL) + goto out; + + if (TC_U32_KEY(ht->handle)) { + u32_remove_hw_knode(tp, (struct tc_u_knode *)ht, extack); + ret = u32_delete_key(tp, (struct tc_u_knode *)ht); + goto out; + } + + if (root_ht == ht) { + NL_SET_ERR_MSG_MOD(extack, "Not allowed to delete root node"); + return -EINVAL; + } + + if (ht->refcnt == 1) { + u32_destroy_hnode(tp, ht, extack); + } else { + NL_SET_ERR_MSG_MOD(extack, "Can not delete in-use filter"); + return -EBUSY; + } + +out: + *last = true; + if (root_ht) { + if (root_ht->refcnt > 2) { + *last = false; + goto ret; + } + if (root_ht->refcnt == 2) { + if (!ht_empty(root_ht)) { + *last = false; + goto ret; + } + } + } + + if (tp_c->refcnt > 1) { + *last = false; + goto ret; + } + + if (tp_c->refcnt == 1) { + struct tc_u_hnode *ht; + + for (ht = rtnl_dereference(tp_c->hlist); + ht; + ht = rtnl_dereference(ht->next)) + if (!ht_empty(ht)) { + *last = false; + break; + } + } + +ret: + return ret; +} + +static u32 gen_new_kid(struct tc_u_hnode *ht, u32 htid) +{ + u32 index = htid | 0x800; + u32 max = htid | 0xFFF; + + if (idr_alloc_u32(&ht->handle_idr, NULL, &index, max, GFP_KERNEL)) { + index = htid + 1; + if (idr_alloc_u32(&ht->handle_idr, NULL, &index, max, + GFP_KERNEL)) + index = max; + } + + return index; +} + +static const struct nla_policy u32_policy[TCA_U32_MAX + 1] = { + [TCA_U32_CLASSID] = { .type = NLA_U32 }, + [TCA_U32_HASH] = { .type = NLA_U32 }, + [TCA_U32_LINK] = { .type = NLA_U32 }, + [TCA_U32_DIVISOR] = { .type = NLA_U32 }, + [TCA_U32_SEL] = { .len = sizeof(struct tc_u32_sel) }, + [TCA_U32_INDEV] = { .type = NLA_STRING, .len = IFNAMSIZ }, + [TCA_U32_MARK] = { .len = sizeof(struct tc_u32_mark) }, + [TCA_U32_FLAGS] = { .type = NLA_U32 }, +}; + +static int u32_set_parms(struct net *net, struct tcf_proto *tp, + unsigned long base, struct tc_u_hnode *ht, + struct tc_u_knode *n, struct nlattr **tb, + struct nlattr *est, bool ovr, + struct netlink_ext_ack *extack) +{ + int err; + + err = tcf_exts_validate(net, tp, tb, est, &n->exts, ovr, extack); + if (err < 0) + return err; + + if (tb[TCA_U32_LINK]) { + u32 handle = nla_get_u32(tb[TCA_U32_LINK]); + struct tc_u_hnode *ht_down = NULL, *ht_old; + + if (TC_U32_KEY(handle)) { + NL_SET_ERR_MSG_MOD(extack, "u32 Link handle must be a hash table"); + return -EINVAL; + } + + if (handle) { + ht_down = u32_lookup_ht(ht->tp_c, handle); + + if (!ht_down) { + NL_SET_ERR_MSG_MOD(extack, "Link hash table not found"); + return -EINVAL; + } + ht_down->refcnt++; + } + + ht_old = rtnl_dereference(n->ht_down); + rcu_assign_pointer(n->ht_down, ht_down); + + if (ht_old) + ht_old->refcnt--; + } + if (tb[TCA_U32_CLASSID]) { + n->res.classid = nla_get_u32(tb[TCA_U32_CLASSID]); + tcf_bind_filter(tp, &n->res, base); + } + +#ifdef CONFIG_NET_CLS_IND + if (tb[TCA_U32_INDEV]) { + int ret; + ret = tcf_change_indev(net, tb[TCA_U32_INDEV], extack); + if (ret < 0) + return -EINVAL; + n->ifindex = ret; + } +#endif + return 0; +} + +static void u32_replace_knode(struct tcf_proto *tp, struct tc_u_common *tp_c, + struct tc_u_knode *n) +{ + struct tc_u_knode __rcu **ins; + struct tc_u_knode *pins; + struct tc_u_hnode *ht; + + if (TC_U32_HTID(n->handle) == TC_U32_ROOT) + ht = rtnl_dereference(tp->root); + else + ht = u32_lookup_ht(tp_c, TC_U32_HTID(n->handle)); + + ins = &ht->ht[TC_U32_HASH(n->handle)]; + + /* The node must always exist for it to be replaced if this is not the + * case then something went very wrong elsewhere. + */ + for (pins = rtnl_dereference(*ins); ; + ins = &pins->next, pins = rtnl_dereference(*ins)) + if (pins->handle == n->handle) + break; + + idr_replace(&ht->handle_idr, n, n->handle); + RCU_INIT_POINTER(n->next, pins->next); + rcu_assign_pointer(*ins, n); +} + +static struct tc_u_knode *u32_init_knode(struct tcf_proto *tp, + struct tc_u_knode *n) +{ + struct tc_u_hnode *ht = rtnl_dereference(n->ht_down); + struct tc_u32_sel *s = &n->sel; + struct tc_u_knode *new; + + new = kzalloc(sizeof(*n) + s->nkeys*sizeof(struct tc_u32_key), + GFP_KERNEL); + + if (!new) + return NULL; + + RCU_INIT_POINTER(new->next, n->next); + new->handle = n->handle; + RCU_INIT_POINTER(new->ht_up, n->ht_up); + +#ifdef CONFIG_NET_CLS_IND + new->ifindex = n->ifindex; +#endif + new->fshift = n->fshift; + new->res = n->res; + new->flags = n->flags; + RCU_INIT_POINTER(new->ht_down, ht); + +#ifdef CONFIG_CLS_U32_PERF + /* Statistics may be incremented by readers during update + * so we must keep them in tact. When the node is later destroyed + * a special destroy call must be made to not free the pf memory. + */ + new->pf = n->pf; +#endif + +#ifdef CONFIG_CLS_U32_MARK + new->val = n->val; + new->mask = n->mask; + /* Similarly success statistics must be moved as pointers */ + new->pcpu_success = n->pcpu_success; +#endif + new->tp = tp; + memcpy(&new->sel, s, sizeof(*s) + s->nkeys*sizeof(struct tc_u32_key)); + + if (tcf_exts_init(&new->exts, TCA_U32_ACT, TCA_U32_POLICE)) { + kfree(new); + return NULL; + } + + /* bump reference count as long as we hold pointer to structure */ + if (ht) + ht->refcnt++; + + return new; +} + +static int u32_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, u32 handle, + struct nlattr **tca, void **arg, bool ovr, + struct netlink_ext_ack *extack) +{ + struct tc_u_common *tp_c = tp->data; + struct tc_u_hnode *ht; + struct tc_u_knode *n; + struct tc_u32_sel *s; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_U32_MAX + 1]; + u32 htid, flags = 0; + size_t sel_size; + int err; +#ifdef CONFIG_CLS_U32_PERF + size_t size; +#endif + + if (!opt) { + if (handle) { + NL_SET_ERR_MSG_MOD(extack, "Filter handle requires options"); + return -EINVAL; + } else { + return 0; + } + } + + err = nla_parse_nested(tb, TCA_U32_MAX, opt, u32_policy, extack); + if (err < 0) + return err; + + if (tb[TCA_U32_FLAGS]) { + flags = nla_get_u32(tb[TCA_U32_FLAGS]); + if (!tc_flags_valid(flags)) { + NL_SET_ERR_MSG_MOD(extack, "Invalid filter flags"); + return -EINVAL; + } + } + + n = *arg; + if (n) { + struct tc_u_knode *new; + + if (TC_U32_KEY(n->handle) == 0) { + NL_SET_ERR_MSG_MOD(extack, "Key node id cannot be zero"); + return -EINVAL; + } + + if ((n->flags ^ flags) & + ~(TCA_CLS_FLAGS_IN_HW | TCA_CLS_FLAGS_NOT_IN_HW)) { + NL_SET_ERR_MSG_MOD(extack, "Key node flags do not match passed flags"); + return -EINVAL; + } + + new = u32_init_knode(tp, n); + if (!new) + return -ENOMEM; + + err = u32_set_parms(net, tp, base, + rtnl_dereference(n->ht_up), new, tb, + tca[TCA_RATE], ovr, extack); + + if (err) { + __u32_destroy_key(new); + return err; + } + + err = u32_replace_hw_knode(tp, new, flags, extack); + if (err) { + __u32_destroy_key(new); + return err; + } + + if (!tc_in_hw(new->flags)) + new->flags |= TCA_CLS_FLAGS_NOT_IN_HW; + + u32_replace_knode(tp, tp_c, new); + tcf_unbind_filter(tp, &n->res); + tcf_exts_get_net(&n->exts); + tcf_queue_work(&n->rwork, u32_delete_key_work); + return 0; + } + + if (tb[TCA_U32_DIVISOR]) { + unsigned int divisor = nla_get_u32(tb[TCA_U32_DIVISOR]); + + if (--divisor > 0x100) { + NL_SET_ERR_MSG_MOD(extack, "Exceeded maximum 256 hash buckets"); + return -EINVAL; + } + if (TC_U32_KEY(handle)) { + NL_SET_ERR_MSG_MOD(extack, "Divisor can only be used on a hash table"); + return -EINVAL; + } + ht = kzalloc(sizeof(*ht) + divisor*sizeof(void *), GFP_KERNEL); + if (ht == NULL) + return -ENOBUFS; + if (handle == 0) { + handle = gen_new_htid(tp->data, ht); + if (handle == 0) { + kfree(ht); + return -ENOMEM; + } + } else { + err = idr_alloc_u32(&tp_c->handle_idr, ht, &handle, + handle, GFP_KERNEL); + if (err) { + kfree(ht); + return err; + } + } + ht->tp_c = tp_c; + ht->refcnt = 1; + ht->divisor = divisor; + ht->handle = handle; + ht->prio = tp->prio; + idr_init(&ht->handle_idr); + ht->flags = flags; + + err = u32_replace_hw_hnode(tp, ht, flags, extack); + if (err) { + idr_remove(&tp_c->handle_idr, handle); + kfree(ht); + return err; + } + + RCU_INIT_POINTER(ht->next, tp_c->hlist); + rcu_assign_pointer(tp_c->hlist, ht); + *arg = ht; + + return 0; + } + + if (tb[TCA_U32_HASH]) { + htid = nla_get_u32(tb[TCA_U32_HASH]); + if (TC_U32_HTID(htid) == TC_U32_ROOT) { + ht = rtnl_dereference(tp->root); + htid = ht->handle; + } else { + ht = u32_lookup_ht(tp->data, TC_U32_HTID(htid)); + if (!ht) { + NL_SET_ERR_MSG_MOD(extack, "Specified hash table not found"); + return -EINVAL; + } + } + } else { + ht = rtnl_dereference(tp->root); + htid = ht->handle; + } + + if (ht->divisor < TC_U32_HASH(htid)) { + NL_SET_ERR_MSG_MOD(extack, "Specified hash table buckets exceed configured value"); + return -EINVAL; + } + + if (handle) { + if (TC_U32_HTID(handle) && TC_U32_HTID(handle ^ htid)) { + NL_SET_ERR_MSG_MOD(extack, "Handle specified hash table address mismatch"); + return -EINVAL; + } + handle = htid | TC_U32_NODE(handle); + err = idr_alloc_u32(&ht->handle_idr, NULL, &handle, handle, + GFP_KERNEL); + if (err) + return err; + } else + handle = gen_new_kid(ht, htid); + + if (tb[TCA_U32_SEL] == NULL) { + NL_SET_ERR_MSG_MOD(extack, "Selector not specified"); + err = -EINVAL; + goto erridr; + } + + s = nla_data(tb[TCA_U32_SEL]); + sel_size = struct_size(s, keys, s->nkeys); + if (nla_len(tb[TCA_U32_SEL]) < sel_size) { + err = -EINVAL; + goto erridr; + } + + n = kzalloc(offsetof(typeof(*n), sel) + sel_size, GFP_KERNEL); + if (n == NULL) { + err = -ENOBUFS; + goto erridr; + } + +#ifdef CONFIG_CLS_U32_PERF + size = sizeof(struct tc_u32_pcnt) + s->nkeys * sizeof(u64); + n->pf = __alloc_percpu(size, __alignof__(struct tc_u32_pcnt)); + if (!n->pf) { + err = -ENOBUFS; + goto errfree; + } +#endif + + memcpy(&n->sel, s, sel_size); + RCU_INIT_POINTER(n->ht_up, ht); + n->handle = handle; + n->fshift = s->hmask ? ffs(ntohl(s->hmask)) - 1 : 0; + n->flags = flags; + n->tp = tp; + + err = tcf_exts_init(&n->exts, TCA_U32_ACT, TCA_U32_POLICE); + if (err < 0) + goto errout; + +#ifdef CONFIG_CLS_U32_MARK + n->pcpu_success = alloc_percpu(u32); + if (!n->pcpu_success) { + err = -ENOMEM; + goto errout; + } + + if (tb[TCA_U32_MARK]) { + struct tc_u32_mark *mark; + + mark = nla_data(tb[TCA_U32_MARK]); + n->val = mark->val; + n->mask = mark->mask; + } +#endif + + err = u32_set_parms(net, tp, base, ht, n, tb, tca[TCA_RATE], ovr, + extack); + if (err == 0) { + struct tc_u_knode __rcu **ins; + struct tc_u_knode *pins; + + err = u32_replace_hw_knode(tp, n, flags, extack); + if (err) + goto errhw; + + if (!tc_in_hw(n->flags)) + n->flags |= TCA_CLS_FLAGS_NOT_IN_HW; + + ins = &ht->ht[TC_U32_HASH(handle)]; + for (pins = rtnl_dereference(*ins); pins; + ins = &pins->next, pins = rtnl_dereference(*ins)) + if (TC_U32_NODE(handle) < TC_U32_NODE(pins->handle)) + break; + + RCU_INIT_POINTER(n->next, pins); + rcu_assign_pointer(*ins, n); + *arg = n; + return 0; + } + +errhw: +#ifdef CONFIG_CLS_U32_MARK + free_percpu(n->pcpu_success); +#endif + +errout: + tcf_exts_destroy(&n->exts); +#ifdef CONFIG_CLS_U32_PERF +errfree: + free_percpu(n->pf); +#endif + kfree(n); +erridr: + idr_remove(&ht->handle_idr, handle); + return err; +} + +static void u32_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct tc_u_common *tp_c = tp->data; + struct tc_u_hnode *ht; + struct tc_u_knode *n; + unsigned int h; + + if (arg->stop) + return; + + for (ht = rtnl_dereference(tp_c->hlist); + ht; + ht = rtnl_dereference(ht->next)) { + if (ht->prio != tp->prio) + continue; + if (arg->count >= arg->skip) { + if (arg->fn(tp, ht, arg) < 0) { + arg->stop = 1; + return; + } + } + arg->count++; + for (h = 0; h <= ht->divisor; h++) { + for (n = rtnl_dereference(ht->ht[h]); + n; + n = rtnl_dereference(n->next)) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(tp, n, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } + } +} + +static int u32_reoffload_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht, + bool add, tc_setup_cb_t *cb, void *cb_priv, + struct netlink_ext_ack *extack) +{ + struct tc_cls_u32_offload cls_u32 = {}; + int err; + + tc_cls_common_offload_init(&cls_u32.common, tp, ht->flags, extack); + cls_u32.command = add ? TC_CLSU32_NEW_HNODE : TC_CLSU32_DELETE_HNODE; + cls_u32.hnode.divisor = ht->divisor; + cls_u32.hnode.handle = ht->handle; + cls_u32.hnode.prio = ht->prio; + + err = cb(TC_SETUP_CLSU32, &cls_u32, cb_priv); + if (err && add && tc_skip_sw(ht->flags)) + return err; + + return 0; +} + +static int u32_reoffload_knode(struct tcf_proto *tp, struct tc_u_knode *n, + bool add, tc_setup_cb_t *cb, void *cb_priv, + struct netlink_ext_ack *extack) +{ + struct tc_u_hnode *ht = rtnl_dereference(n->ht_down); + struct tcf_block *block = tp->chain->block; + struct tc_cls_u32_offload cls_u32 = {}; + int err; + + tc_cls_common_offload_init(&cls_u32.common, tp, n->flags, extack); + cls_u32.command = add ? + TC_CLSU32_REPLACE_KNODE : TC_CLSU32_DELETE_KNODE; + cls_u32.knode.handle = n->handle; + + if (add) { + cls_u32.knode.fshift = n->fshift; +#ifdef CONFIG_CLS_U32_MARK + cls_u32.knode.val = n->val; + cls_u32.knode.mask = n->mask; +#else + cls_u32.knode.val = 0; + cls_u32.knode.mask = 0; +#endif + cls_u32.knode.sel = &n->sel; + cls_u32.knode.exts = &n->exts; + if (n->ht_down) + cls_u32.knode.link_handle = ht->handle; + } + + err = cb(TC_SETUP_CLSU32, &cls_u32, cb_priv); + if (err) { + if (add && tc_skip_sw(n->flags)) + return err; + return 0; + } + + tc_cls_offload_cnt_update(block, &n->in_hw_count, &n->flags, add); + + return 0; +} + +static int u32_reoffload(struct tcf_proto *tp, bool add, tc_setup_cb_t *cb, + void *cb_priv, struct netlink_ext_ack *extack) +{ + struct tc_u_common *tp_c = tp->data; + struct tc_u_hnode *ht; + struct tc_u_knode *n; + unsigned int h; + int err; + + for (ht = rtnl_dereference(tp_c->hlist); + ht; + ht = rtnl_dereference(ht->next)) { + if (ht->prio != tp->prio) + continue; + + /* When adding filters to a new dev, try to offload the + * hashtable first. When removing, do the filters before the + * hashtable. + */ + if (add && !tc_skip_hw(ht->flags)) { + err = u32_reoffload_hnode(tp, ht, add, cb, cb_priv, + extack); + if (err) + return err; + } + + for (h = 0; h <= ht->divisor; h++) { + for (n = rtnl_dereference(ht->ht[h]); + n; + n = rtnl_dereference(n->next)) { + if (tc_skip_hw(n->flags)) + continue; + + err = u32_reoffload_knode(tp, n, add, cb, + cb_priv, extack); + if (err) + return err; + } + } + + if (!add && !tc_skip_hw(ht->flags)) + u32_reoffload_hnode(tp, ht, add, cb, cb_priv, extack); + } + + return 0; +} + +static void u32_bind_class(void *fh, u32 classid, unsigned long cl, void *q, + unsigned long base) +{ + struct tc_u_knode *n = fh; + + if (n && n->res.classid == classid) { + if (cl) + __tcf_bind_filter(q, &n->res, base); + else + __tcf_unbind_filter(q, &n->res); + } +} + +static int u32_dump(struct net *net, struct tcf_proto *tp, void *fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct tc_u_knode *n = fh; + struct tc_u_hnode *ht_up, *ht_down; + struct nlattr *nest; + + if (n == NULL) + return skb->len; + + t->tcm_handle = n->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (TC_U32_KEY(n->handle) == 0) { + struct tc_u_hnode *ht = fh; + u32 divisor = ht->divisor + 1; + + if (nla_put_u32(skb, TCA_U32_DIVISOR, divisor)) + goto nla_put_failure; + } else { +#ifdef CONFIG_CLS_U32_PERF + struct tc_u32_pcnt *gpf; + int cpu; +#endif + + if (nla_put(skb, TCA_U32_SEL, + sizeof(n->sel) + n->sel.nkeys*sizeof(struct tc_u32_key), + &n->sel)) + goto nla_put_failure; + + ht_up = rtnl_dereference(n->ht_up); + if (ht_up) { + u32 htid = n->handle & 0xFFFFF000; + if (nla_put_u32(skb, TCA_U32_HASH, htid)) + goto nla_put_failure; + } + if (n->res.classid && + nla_put_u32(skb, TCA_U32_CLASSID, n->res.classid)) + goto nla_put_failure; + + ht_down = rtnl_dereference(n->ht_down); + if (ht_down && + nla_put_u32(skb, TCA_U32_LINK, ht_down->handle)) + goto nla_put_failure; + + if (n->flags && nla_put_u32(skb, TCA_U32_FLAGS, n->flags)) + goto nla_put_failure; + +#ifdef CONFIG_CLS_U32_MARK + if ((n->val || n->mask)) { + struct tc_u32_mark mark = {.val = n->val, + .mask = n->mask, + .success = 0}; + int cpum; + + for_each_possible_cpu(cpum) { + __u32 cnt = *per_cpu_ptr(n->pcpu_success, cpum); + + mark.success += cnt; + } + + if (nla_put(skb, TCA_U32_MARK, sizeof(mark), &mark)) + goto nla_put_failure; + } +#endif + + if (tcf_exts_dump(skb, &n->exts) < 0) + goto nla_put_failure; + +#ifdef CONFIG_NET_CLS_IND + if (n->ifindex) { + struct net_device *dev; + dev = __dev_get_by_index(net, n->ifindex); + if (dev && nla_put_string(skb, TCA_U32_INDEV, dev->name)) + goto nla_put_failure; + } +#endif +#ifdef CONFIG_CLS_U32_PERF + gpf = kzalloc(sizeof(struct tc_u32_pcnt) + + n->sel.nkeys * sizeof(u64), + GFP_KERNEL); + if (!gpf) + goto nla_put_failure; + + for_each_possible_cpu(cpu) { + int i; + struct tc_u32_pcnt *pf = per_cpu_ptr(n->pf, cpu); + + gpf->rcnt += pf->rcnt; + gpf->rhit += pf->rhit; + for (i = 0; i < n->sel.nkeys; i++) + gpf->kcnts[i] += pf->kcnts[i]; + } + + if (nla_put_64bit(skb, TCA_U32_PCNT, + sizeof(struct tc_u32_pcnt) + + n->sel.nkeys * sizeof(u64), + gpf, TCA_U32_PAD)) { + kfree(gpf); + goto nla_put_failure; + } + kfree(gpf); +#endif + } + + nla_nest_end(skb, nest); + + if (TC_U32_KEY(n->handle)) + if (tcf_exts_dump_stats(skb, &n->exts) < 0) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static struct tcf_proto_ops cls_u32_ops __read_mostly = { + .kind = "u32", + .classify = u32_classify, + .init = u32_init, + .destroy = u32_destroy, + .get = u32_get, + .change = u32_change, + .delete = u32_delete, + .walk = u32_walk, + .reoffload = u32_reoffload, + .dump = u32_dump, + .bind_class = u32_bind_class, + .owner = THIS_MODULE, +}; + +static int __init init_u32(void) +{ + int i, ret; + + pr_info("u32 classifier\n"); +#ifdef CONFIG_CLS_U32_PERF + pr_info(" Performance counters on\n"); +#endif +#ifdef CONFIG_NET_CLS_IND + pr_info(" input device check on\n"); +#endif +#ifdef CONFIG_NET_CLS_ACT + pr_info(" Actions configured\n"); +#endif + tc_u_common_hash = kvmalloc_array(U32_HASH_SIZE, + sizeof(struct hlist_head), + GFP_KERNEL); + if (!tc_u_common_hash) + return -ENOMEM; + + for (i = 0; i < U32_HASH_SIZE; i++) + INIT_HLIST_HEAD(&tc_u_common_hash[i]); + + ret = register_tcf_proto_ops(&cls_u32_ops); + if (ret) + kvfree(tc_u_common_hash); + return ret; +} + +static void __exit exit_u32(void) +{ + unregister_tcf_proto_ops(&cls_u32_ops); + kvfree(tc_u_common_hash); +} + +module_init(init_u32) +module_exit(exit_u32) +MODULE_LICENSE("GPL"); diff --git a/net/sched/em_canid.c b/net/sched/em_canid.c new file mode 100644 index 000000000..ddd883ca5 --- /dev/null +++ b/net/sched/em_canid.c @@ -0,0 +1,233 @@ +/* + * em_canid.c Ematch rule to match CAN frames according to their CAN IDs + * + * This program is free software; you can distribute 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. + * + * Idea: Oliver Hartkopp <oliver.hartkopp@volkswagen.de> + * Copyright: (c) 2011 Czech Technical University in Prague + * (c) 2011 Volkswagen Group Research + * Authors: Michal Sojka <sojkam1@fel.cvut.cz> + * Pavel Pisa <pisa@cmp.felk.cvut.cz> + * Rostislav Lisovy <lisovy@gmail.cz> + * Funded by: Volkswagen Group Research + */ + +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <net/pkt_cls.h> +#include <linux/can.h> + +#define EM_CAN_RULES_MAX 500 + +struct canid_match { + /* For each SFF CAN ID (11 bit) there is one record in this bitfield */ + DECLARE_BITMAP(match_sff, (1 << CAN_SFF_ID_BITS)); + + int rules_count; + int sff_rules_count; + int eff_rules_count; + + /* + * Raw rules copied from netlink message; Used for sending + * information to userspace (when 'tc filter show' is invoked) + * AND when matching EFF frames + */ + struct can_filter rules_raw[]; +}; + +/** + * em_canid_get_id() - Extracts Can ID out of the sk_buff structure. + */ +static canid_t em_canid_get_id(struct sk_buff *skb) +{ + /* CAN ID is stored within the data field */ + struct can_frame *cf = (struct can_frame *)skb->data; + + return cf->can_id; +} + +static void em_canid_sff_match_add(struct canid_match *cm, u32 can_id, + u32 can_mask) +{ + int i; + + /* + * Limit can_mask and can_id to SFF range to + * protect against write after end of array + */ + can_mask &= CAN_SFF_MASK; + can_id &= can_mask; + + /* Single frame */ + if (can_mask == CAN_SFF_MASK) { + set_bit(can_id, cm->match_sff); + return; + } + + /* All frames */ + if (can_mask == 0) { + bitmap_fill(cm->match_sff, (1 << CAN_SFF_ID_BITS)); + return; + } + + /* + * Individual frame filter. + * Add record (set bit to 1) for each ID that + * conforms particular rule + */ + for (i = 0; i < (1 << CAN_SFF_ID_BITS); i++) { + if ((i & can_mask) == can_id) + set_bit(i, cm->match_sff); + } +} + +static inline struct canid_match *em_canid_priv(struct tcf_ematch *m) +{ + return (struct canid_match *)m->data; +} + +static int em_canid_match(struct sk_buff *skb, struct tcf_ematch *m, + struct tcf_pkt_info *info) +{ + struct canid_match *cm = em_canid_priv(m); + canid_t can_id; + int match = 0; + int i; + const struct can_filter *lp; + + can_id = em_canid_get_id(skb); + + if (can_id & CAN_EFF_FLAG) { + for (i = 0, lp = cm->rules_raw; + i < cm->eff_rules_count; i++, lp++) { + if (!(((lp->can_id ^ can_id) & lp->can_mask))) { + match = 1; + break; + } + } + } else { /* SFF */ + can_id &= CAN_SFF_MASK; + match = (test_bit(can_id, cm->match_sff) ? 1 : 0); + } + + return match; +} + +static int em_canid_change(struct net *net, void *data, int len, + struct tcf_ematch *m) +{ + struct can_filter *conf = data; /* Array with rules */ + struct canid_match *cm; + int i; + + if (!len) + return -EINVAL; + + if (len % sizeof(struct can_filter)) + return -EINVAL; + + if (len > sizeof(struct can_filter) * EM_CAN_RULES_MAX) + return -EINVAL; + + cm = kzalloc(sizeof(struct canid_match) + len, GFP_KERNEL); + if (!cm) + return -ENOMEM; + + cm->rules_count = len / sizeof(struct can_filter); + + /* + * We need two for() loops for copying rules into two contiguous + * areas in rules_raw to process all eff rules with a simple loop. + * NB: The configuration interface supports sff and eff rules. + * We do not support filters here that match for the same can_id + * provided in a SFF and EFF frame (e.g. 0x123 / 0x80000123). + * For this (unusual case) two filters have to be specified. The + * SFF/EFF separation is done with the CAN_EFF_FLAG in the can_id. + */ + + /* Fill rules_raw with EFF rules first */ + for (i = 0; i < cm->rules_count; i++) { + if (conf[i].can_id & CAN_EFF_FLAG) { + memcpy(cm->rules_raw + cm->eff_rules_count, + &conf[i], + sizeof(struct can_filter)); + + cm->eff_rules_count++; + } + } + + /* append SFF frame rules */ + for (i = 0; i < cm->rules_count; i++) { + if (!(conf[i].can_id & CAN_EFF_FLAG)) { + memcpy(cm->rules_raw + + cm->eff_rules_count + + cm->sff_rules_count, + &conf[i], sizeof(struct can_filter)); + + cm->sff_rules_count++; + + em_canid_sff_match_add(cm, + conf[i].can_id, conf[i].can_mask); + } + } + + m->datalen = sizeof(struct canid_match) + len; + m->data = (unsigned long)cm; + return 0; +} + +static void em_canid_destroy(struct tcf_ematch *m) +{ + struct canid_match *cm = em_canid_priv(m); + + kfree(cm); +} + +static int em_canid_dump(struct sk_buff *skb, struct tcf_ematch *m) +{ + struct canid_match *cm = em_canid_priv(m); + + /* + * When configuring this ematch 'rules_count' is set not to exceed + * 'rules_raw' array size + */ + if (nla_put_nohdr(skb, sizeof(struct can_filter) * cm->rules_count, + &cm->rules_raw) < 0) + return -EMSGSIZE; + + return 0; +} + +static struct tcf_ematch_ops em_canid_ops = { + .kind = TCF_EM_CANID, + .change = em_canid_change, + .match = em_canid_match, + .destroy = em_canid_destroy, + .dump = em_canid_dump, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_canid_ops.link) +}; + +static int __init init_em_canid(void) +{ + return tcf_em_register(&em_canid_ops); +} + +static void __exit exit_em_canid(void) +{ + tcf_em_unregister(&em_canid_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_canid); +module_exit(exit_em_canid); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_CANID); diff --git a/net/sched/em_cmp.c b/net/sched/em_cmp.c new file mode 100644 index 000000000..1c8360a27 --- /dev/null +++ b/net/sched/em_cmp.c @@ -0,0 +1,99 @@ +/* + * net/sched/em_cmp.c Simple packet data comparison ematch + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/tc_ematch/tc_em_cmp.h> +#include <asm/unaligned.h> +#include <net/pkt_cls.h> + +static inline int cmp_needs_transformation(struct tcf_em_cmp *cmp) +{ + return unlikely(cmp->flags & TCF_EM_CMP_TRANS); +} + +static int em_cmp_match(struct sk_buff *skb, struct tcf_ematch *em, + struct tcf_pkt_info *info) +{ + struct tcf_em_cmp *cmp = (struct tcf_em_cmp *) em->data; + unsigned char *ptr = tcf_get_base_ptr(skb, cmp->layer) + cmp->off; + u32 val = 0; + + if (!tcf_valid_offset(skb, ptr, cmp->align)) + return 0; + + switch (cmp->align) { + case TCF_EM_ALIGN_U8: + val = *ptr; + break; + + case TCF_EM_ALIGN_U16: + val = get_unaligned_be16(ptr); + + if (cmp_needs_transformation(cmp)) + val = be16_to_cpu(val); + break; + + case TCF_EM_ALIGN_U32: + /* Worth checking boundries? The branching seems + * to get worse. Visit again. + */ + val = get_unaligned_be32(ptr); + + if (cmp_needs_transformation(cmp)) + val = be32_to_cpu(val); + break; + + default: + return 0; + } + + if (cmp->mask) + val &= cmp->mask; + + switch (cmp->opnd) { + case TCF_EM_OPND_EQ: + return val == cmp->val; + case TCF_EM_OPND_LT: + return val < cmp->val; + case TCF_EM_OPND_GT: + return val > cmp->val; + } + + return 0; +} + +static struct tcf_ematch_ops em_cmp_ops = { + .kind = TCF_EM_CMP, + .datalen = sizeof(struct tcf_em_cmp), + .match = em_cmp_match, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_cmp_ops.link) +}; + +static int __init init_em_cmp(void) +{ + return tcf_em_register(&em_cmp_ops); +} + +static void __exit exit_em_cmp(void) +{ + tcf_em_unregister(&em_cmp_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_cmp); +module_exit(exit_em_cmp); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_CMP); diff --git a/net/sched/em_ipset.c b/net/sched/em_ipset.c new file mode 100644 index 000000000..ef3b6b66c --- /dev/null +++ b/net/sched/em_ipset.c @@ -0,0 +1,137 @@ +/* + * net/sched/em_ipset.c ipset ematch + * + * Copyright (c) 2012 Florian Westphal <fw@strlen.de> + * + * 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/gfp.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <linux/netfilter/xt_set.h> +#include <linux/ipv6.h> +#include <net/ip.h> +#include <net/pkt_cls.h> + +static int em_ipset_change(struct net *net, void *data, int data_len, + struct tcf_ematch *em) +{ + struct xt_set_info *set = data; + ip_set_id_t index; + + if (data_len != sizeof(*set)) + return -EINVAL; + + index = ip_set_nfnl_get_byindex(net, set->index); + if (index == IPSET_INVALID_ID) + return -ENOENT; + + em->datalen = sizeof(*set); + em->data = (unsigned long)kmemdup(data, em->datalen, GFP_KERNEL); + if (em->data) + return 0; + + ip_set_nfnl_put(net, index); + return -ENOMEM; +} + +static void em_ipset_destroy(struct tcf_ematch *em) +{ + const struct xt_set_info *set = (const void *) em->data; + if (set) { + ip_set_nfnl_put(em->net, set->index); + kfree((void *) em->data); + } +} + +static int em_ipset_match(struct sk_buff *skb, struct tcf_ematch *em, + struct tcf_pkt_info *info) +{ + struct ip_set_adt_opt opt; + struct xt_action_param acpar; + const struct xt_set_info *set = (const void *) em->data; + struct net_device *dev, *indev = NULL; + struct nf_hook_state state = { + .net = em->net, + }; + int ret, network_offset; + + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + state.pf = NFPROTO_IPV4; + if (!pskb_network_may_pull(skb, sizeof(struct iphdr))) + return 0; + acpar.thoff = ip_hdrlen(skb); + break; + case htons(ETH_P_IPV6): + state.pf = NFPROTO_IPV6; + if (!pskb_network_may_pull(skb, sizeof(struct ipv6hdr))) + return 0; + /* doesn't call ipv6_find_hdr() because ipset doesn't use thoff, yet */ + acpar.thoff = sizeof(struct ipv6hdr); + break; + default: + return 0; + } + + opt.family = state.pf; + opt.dim = set->dim; + opt.flags = set->flags; + opt.cmdflags = 0; + opt.ext.timeout = ~0u; + + network_offset = skb_network_offset(skb); + skb_pull(skb, network_offset); + + dev = skb->dev; + + rcu_read_lock(); + + if (skb->skb_iif) + indev = dev_get_by_index_rcu(em->net, skb->skb_iif); + + state.in = indev ? indev : dev; + state.out = dev; + acpar.state = &state; + + ret = ip_set_test(set->index, skb, &acpar, &opt); + + rcu_read_unlock(); + + skb_push(skb, network_offset); + return ret; +} + +static struct tcf_ematch_ops em_ipset_ops = { + .kind = TCF_EM_IPSET, + .change = em_ipset_change, + .destroy = em_ipset_destroy, + .match = em_ipset_match, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_ipset_ops.link) +}; + +static int __init init_em_ipset(void) +{ + return tcf_em_register(&em_ipset_ops); +} + +static void __exit exit_em_ipset(void) +{ + tcf_em_unregister(&em_ipset_ops); +} + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Florian Westphal <fw@strlen.de>"); +MODULE_DESCRIPTION("TC extended match for IP sets"); + +module_init(init_em_ipset); +module_exit(exit_em_ipset); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_IPSET); diff --git a/net/sched/em_ipt.c b/net/sched/em_ipt.c new file mode 100644 index 000000000..a5f34e930 --- /dev/null +++ b/net/sched/em_ipt.c @@ -0,0 +1,257 @@ +/* + * net/sched/em_ipt.c IPtables matches Ematch + * + * (c) 2018 Eyal Birger <eyal.birger@gmail.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/gfp.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <linux/tc_ematch/tc_em_ipt.h> +#include <linux/netfilter.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter_ipv6/ip6_tables.h> +#include <net/pkt_cls.h> + +struct em_ipt_match { + const struct xt_match *match; + u32 hook; + u8 match_data[0] __aligned(8); +}; + +struct em_ipt_xt_match { + char *match_name; + int (*validate_match_data)(struct nlattr **tb, u8 mrev); +}; + +static const struct nla_policy em_ipt_policy[TCA_EM_IPT_MAX + 1] = { + [TCA_EM_IPT_MATCH_NAME] = { .type = NLA_STRING, + .len = XT_EXTENSION_MAXNAMELEN }, + [TCA_EM_IPT_MATCH_REVISION] = { .type = NLA_U8 }, + [TCA_EM_IPT_HOOK] = { .type = NLA_U32 }, + [TCA_EM_IPT_NFPROTO] = { .type = NLA_U8 }, + [TCA_EM_IPT_MATCH_DATA] = { .type = NLA_UNSPEC }, +}; + +static int check_match(struct net *net, struct em_ipt_match *im, int mdata_len) +{ + struct xt_mtchk_param mtpar = {}; + union { + struct ipt_entry e4; + struct ip6t_entry e6; + } e = {}; + + mtpar.net = net; + mtpar.table = "filter"; + mtpar.hook_mask = 1 << im->hook; + mtpar.family = im->match->family; + mtpar.match = im->match; + mtpar.entryinfo = &e; + mtpar.matchinfo = (void *)im->match_data; + return xt_check_match(&mtpar, mdata_len, 0, 0); +} + +static int policy_validate_match_data(struct nlattr **tb, u8 mrev) +{ + if (mrev != 0) { + pr_err("only policy match revision 0 supported"); + return -EINVAL; + } + + if (nla_get_u32(tb[TCA_EM_IPT_HOOK]) != NF_INET_PRE_ROUTING) { + pr_err("policy can only be matched on NF_INET_PRE_ROUTING"); + return -EINVAL; + } + + return 0; +} + +static const struct em_ipt_xt_match em_ipt_xt_matches[] = { + { + .match_name = "policy", + .validate_match_data = policy_validate_match_data + }, + {} +}; + +static struct xt_match *get_xt_match(struct nlattr **tb) +{ + const struct em_ipt_xt_match *m; + struct nlattr *mname_attr; + u8 nfproto, mrev = 0; + int ret; + + mname_attr = tb[TCA_EM_IPT_MATCH_NAME]; + for (m = em_ipt_xt_matches; m->match_name; m++) { + if (!nla_strcmp(mname_attr, m->match_name)) + break; + } + + if (!m->match_name) { + pr_err("Unsupported xt match"); + return ERR_PTR(-EINVAL); + } + + if (tb[TCA_EM_IPT_MATCH_REVISION]) + mrev = nla_get_u8(tb[TCA_EM_IPT_MATCH_REVISION]); + + ret = m->validate_match_data(tb, mrev); + if (ret < 0) + return ERR_PTR(ret); + + nfproto = nla_get_u8(tb[TCA_EM_IPT_NFPROTO]); + return xt_request_find_match(nfproto, m->match_name, mrev); +} + +static int em_ipt_change(struct net *net, void *data, int data_len, + struct tcf_ematch *em) +{ + struct nlattr *tb[TCA_EM_IPT_MAX + 1]; + struct em_ipt_match *im = NULL; + struct xt_match *match; + int mdata_len, ret; + + ret = nla_parse(tb, TCA_EM_IPT_MAX, data, data_len, em_ipt_policy, + NULL); + if (ret < 0) + return ret; + + if (!tb[TCA_EM_IPT_HOOK] || !tb[TCA_EM_IPT_MATCH_NAME] || + !tb[TCA_EM_IPT_MATCH_DATA] || !tb[TCA_EM_IPT_NFPROTO]) + return -EINVAL; + + match = get_xt_match(tb); + if (IS_ERR(match)) { + pr_err("unable to load match\n"); + return PTR_ERR(match); + } + + mdata_len = XT_ALIGN(nla_len(tb[TCA_EM_IPT_MATCH_DATA])); + im = kzalloc(sizeof(*im) + mdata_len, GFP_KERNEL); + if (!im) { + ret = -ENOMEM; + goto err; + } + + im->match = match; + im->hook = nla_get_u32(tb[TCA_EM_IPT_HOOK]); + nla_memcpy(im->match_data, tb[TCA_EM_IPT_MATCH_DATA], mdata_len); + + ret = check_match(net, im, mdata_len); + if (ret) + goto err; + + em->datalen = sizeof(*im) + mdata_len; + em->data = (unsigned long)im; + return 0; + +err: + kfree(im); + module_put(match->me); + return ret; +} + +static void em_ipt_destroy(struct tcf_ematch *em) +{ + struct em_ipt_match *im = (void *)em->data; + + if (!im) + return; + + if (im->match->destroy) { + struct xt_mtdtor_param par = { + .net = em->net, + .match = im->match, + .matchinfo = im->match_data, + .family = im->match->family + }; + im->match->destroy(&par); + } + module_put(im->match->me); + kfree((void *)im); +} + +static int em_ipt_match(struct sk_buff *skb, struct tcf_ematch *em, + struct tcf_pkt_info *info) +{ + const struct em_ipt_match *im = (const void *)em->data; + struct xt_action_param acpar = {}; + struct net_device *indev = NULL; + struct nf_hook_state state; + int ret; + + rcu_read_lock(); + + if (skb->skb_iif) + indev = dev_get_by_index_rcu(em->net, skb->skb_iif); + + nf_hook_state_init(&state, im->hook, im->match->family, + indev ?: skb->dev, skb->dev, NULL, em->net, NULL); + + acpar.match = im->match; + acpar.matchinfo = im->match_data; + acpar.state = &state; + + ret = im->match->match(skb, &acpar); + + rcu_read_unlock(); + return ret; +} + +static int em_ipt_dump(struct sk_buff *skb, struct tcf_ematch *em) +{ + struct em_ipt_match *im = (void *)em->data; + + if (nla_put_string(skb, TCA_EM_IPT_MATCH_NAME, im->match->name) < 0) + return -EMSGSIZE; + if (nla_put_u32(skb, TCA_EM_IPT_HOOK, im->hook) < 0) + return -EMSGSIZE; + if (nla_put_u8(skb, TCA_EM_IPT_MATCH_REVISION, im->match->revision) < 0) + return -EMSGSIZE; + if (nla_put_u8(skb, TCA_EM_IPT_NFPROTO, im->match->family) < 0) + return -EMSGSIZE; + if (nla_put(skb, TCA_EM_IPT_MATCH_DATA, + im->match->usersize ?: im->match->matchsize, + im->match_data) < 0) + return -EMSGSIZE; + + return 0; +} + +static struct tcf_ematch_ops em_ipt_ops = { + .kind = TCF_EM_IPT, + .change = em_ipt_change, + .destroy = em_ipt_destroy, + .match = em_ipt_match, + .dump = em_ipt_dump, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_ipt_ops.link) +}; + +static int __init init_em_ipt(void) +{ + return tcf_em_register(&em_ipt_ops); +} + +static void __exit exit_em_ipt(void) +{ + tcf_em_unregister(&em_ipt_ops); +} + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Eyal Birger <eyal.birger@gmail.com>"); +MODULE_DESCRIPTION("TC extended match for IPtables matches"); + +module_init(init_em_ipt); +module_exit(exit_em_ipt); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_IPT); diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c new file mode 100644 index 000000000..e36fa9272 --- /dev/null +++ b/net/sched/em_meta.c @@ -0,0 +1,1014 @@ +/* + * net/sched/em_meta.c Metadata ematch + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + * + * ========================================================================== + * + * The metadata ematch compares two meta objects where each object + * represents either a meta value stored in the kernel or a static + * value provided by userspace. The objects are not provided by + * userspace itself but rather a definition providing the information + * to build them. Every object is of a certain type which must be + * equal to the object it is being compared to. + * + * The definition of a objects conists of the type (meta type), a + * identifier (meta id) and additional type specific information. + * The meta id is either TCF_META_TYPE_VALUE for values provided by + * userspace or a index to the meta operations table consisting of + * function pointers to type specific meta data collectors returning + * the value of the requested meta value. + * + * lvalue rvalue + * +-----------+ +-----------+ + * | type: INT | | type: INT | + * def | id: DEV | | id: VALUE | + * | data: | | data: 3 | + * +-----------+ +-----------+ + * | | + * ---> meta_ops[INT][DEV](...) | + * | | + * ----------- | + * V V + * +-----------+ +-----------+ + * | type: INT | | type: INT | + * obj | id: DEV | | id: VALUE | + * | data: 2 |<--data got filled out | data: 3 | + * +-----------+ +-----------+ + * | | + * --------------> 2 equals 3 <-------------- + * + * This is a simplified schema, the complexity varies depending + * on the meta type. Obviously, the length of the data must also + * be provided for non-numeric types. + * + * Additionally, type dependent modifiers such as shift operators + * or mask may be applied to extend the functionaliy. As of now, + * the variable length type supports shifting the byte string to + * the right, eating up any number of octets and thus supporting + * wildcard interface name comparisons such as "ppp%" matching + * ppp0..9. + * + * NOTE: Certain meta values depend on other subsystems and are + * only available if that subsystem is enabled in the kernel. + */ + +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/sched/loadavg.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <linux/random.h> +#include <linux/if_vlan.h> +#include <linux/tc_ematch/tc_em_meta.h> +#include <net/dst.h> +#include <net/route.h> +#include <net/pkt_cls.h> +#include <net/sock.h> + +struct meta_obj { + unsigned long value; + unsigned int len; +}; + +struct meta_value { + struct tcf_meta_val hdr; + unsigned long val; + unsigned int len; +}; + +struct meta_match { + struct meta_value lvalue; + struct meta_value rvalue; +}; + +static inline int meta_id(struct meta_value *v) +{ + return TCF_META_ID(v->hdr.kind); +} + +static inline int meta_type(struct meta_value *v) +{ + return TCF_META_TYPE(v->hdr.kind); +} + +#define META_COLLECTOR(FUNC) static void meta_##FUNC(struct sk_buff *skb, \ + struct tcf_pkt_info *info, struct meta_value *v, \ + struct meta_obj *dst, int *err) + +/************************************************************************** + * System status & misc + **************************************************************************/ + +META_COLLECTOR(int_random) +{ + get_random_bytes(&dst->value, sizeof(dst->value)); +} + +static inline unsigned long fixed_loadavg(int load) +{ + int rnd_load = load + (FIXED_1/200); + int rnd_frac = ((rnd_load & (FIXED_1-1)) * 100) >> FSHIFT; + + return ((rnd_load >> FSHIFT) * 100) + rnd_frac; +} + +META_COLLECTOR(int_loadavg_0) +{ + dst->value = fixed_loadavg(avenrun[0]); +} + +META_COLLECTOR(int_loadavg_1) +{ + dst->value = fixed_loadavg(avenrun[1]); +} + +META_COLLECTOR(int_loadavg_2) +{ + dst->value = fixed_loadavg(avenrun[2]); +} + +/************************************************************************** + * Device names & indices + **************************************************************************/ + +static inline int int_dev(struct net_device *dev, struct meta_obj *dst) +{ + if (unlikely(dev == NULL)) + return -1; + + dst->value = dev->ifindex; + return 0; +} + +static inline int var_dev(struct net_device *dev, struct meta_obj *dst) +{ + if (unlikely(dev == NULL)) + return -1; + + dst->value = (unsigned long) dev->name; + dst->len = strlen(dev->name); + return 0; +} + +META_COLLECTOR(int_dev) +{ + *err = int_dev(skb->dev, dst); +} + +META_COLLECTOR(var_dev) +{ + *err = var_dev(skb->dev, dst); +} + +/************************************************************************** + * vlan tag + **************************************************************************/ + +META_COLLECTOR(int_vlan_tag) +{ + unsigned short tag; + + if (skb_vlan_tag_present(skb)) + dst->value = skb_vlan_tag_get(skb); + else if (!__vlan_get_tag(skb, &tag)) + dst->value = tag; + else + *err = -1; +} + + + +/************************************************************************** + * skb attributes + **************************************************************************/ + +META_COLLECTOR(int_priority) +{ + dst->value = skb->priority; +} + +META_COLLECTOR(int_protocol) +{ + /* Let userspace take care of the byte ordering */ + dst->value = skb_protocol(skb, false); +} + +META_COLLECTOR(int_pkttype) +{ + dst->value = skb->pkt_type; +} + +META_COLLECTOR(int_pktlen) +{ + dst->value = skb->len; +} + +META_COLLECTOR(int_datalen) +{ + dst->value = skb->data_len; +} + +META_COLLECTOR(int_maclen) +{ + dst->value = skb->mac_len; +} + +META_COLLECTOR(int_rxhash) +{ + dst->value = skb_get_hash(skb); +} + +/************************************************************************** + * Netfilter + **************************************************************************/ + +META_COLLECTOR(int_mark) +{ + dst->value = skb->mark; +} + +/************************************************************************** + * Traffic Control + **************************************************************************/ + +META_COLLECTOR(int_tcindex) +{ + dst->value = skb->tc_index; +} + +/************************************************************************** + * Routing + **************************************************************************/ + +META_COLLECTOR(int_rtclassid) +{ + if (unlikely(skb_dst(skb) == NULL)) + *err = -1; + else +#ifdef CONFIG_IP_ROUTE_CLASSID + dst->value = skb_dst(skb)->tclassid; +#else + dst->value = 0; +#endif +} + +META_COLLECTOR(int_rtiif) +{ + if (unlikely(skb_rtable(skb) == NULL)) + *err = -1; + else + dst->value = inet_iif(skb); +} + +/************************************************************************** + * Socket Attributes + **************************************************************************/ + +#define skip_nonlocal(skb) \ + (unlikely(skb->sk == NULL)) + +META_COLLECTOR(int_sk_family) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + dst->value = skb->sk->sk_family; +} + +META_COLLECTOR(int_sk_state) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + dst->value = skb->sk->sk_state; +} + +META_COLLECTOR(int_sk_reuse) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + dst->value = skb->sk->sk_reuse; +} + +META_COLLECTOR(int_sk_bound_if) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + /* No error if bound_dev_if is 0, legal userspace check */ + dst->value = skb->sk->sk_bound_dev_if; +} + +META_COLLECTOR(var_sk_bound_if) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + + if (skb->sk->sk_bound_dev_if == 0) { + dst->value = (unsigned long) "any"; + dst->len = 3; + } else { + struct net_device *dev; + + rcu_read_lock(); + dev = dev_get_by_index_rcu(sock_net(skb->sk), + skb->sk->sk_bound_dev_if); + *err = var_dev(dev, dst); + rcu_read_unlock(); + } +} + +META_COLLECTOR(int_sk_refcnt) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + dst->value = refcount_read(&skb->sk->sk_refcnt); +} + +META_COLLECTOR(int_sk_rcvbuf) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_rcvbuf; +} + +META_COLLECTOR(int_sk_shutdown) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_shutdown; +} + +META_COLLECTOR(int_sk_proto) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_protocol; +} + +META_COLLECTOR(int_sk_type) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_type; +} + +META_COLLECTOR(int_sk_rmem_alloc) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk_rmem_alloc_get(sk); +} + +META_COLLECTOR(int_sk_wmem_alloc) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk_wmem_alloc_get(sk); +} + +META_COLLECTOR(int_sk_omem_alloc) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = atomic_read(&sk->sk_omem_alloc); +} + +META_COLLECTOR(int_sk_rcv_qlen) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_receive_queue.qlen; +} + +META_COLLECTOR(int_sk_snd_qlen) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_write_queue.qlen; +} + +META_COLLECTOR(int_sk_wmem_queued) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_wmem_queued; +} + +META_COLLECTOR(int_sk_fwd_alloc) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_forward_alloc; +} + +META_COLLECTOR(int_sk_sndbuf) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_sndbuf; +} + +META_COLLECTOR(int_sk_alloc) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = (__force int) sk->sk_allocation; +} + +META_COLLECTOR(int_sk_hash) +{ + if (skip_nonlocal(skb)) { + *err = -1; + return; + } + dst->value = skb->sk->sk_hash; +} + +META_COLLECTOR(int_sk_lingertime) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_lingertime / HZ; +} + +META_COLLECTOR(int_sk_err_qlen) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_error_queue.qlen; +} + +META_COLLECTOR(int_sk_ack_bl) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_ack_backlog; +} + +META_COLLECTOR(int_sk_max_ack_bl) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_max_ack_backlog; +} + +META_COLLECTOR(int_sk_prio) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_priority; +} + +META_COLLECTOR(int_sk_rcvlowat) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_rcvlowat; +} + +META_COLLECTOR(int_sk_rcvtimeo) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_rcvtimeo / HZ; +} + +META_COLLECTOR(int_sk_sndtimeo) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_sndtimeo / HZ; +} + +META_COLLECTOR(int_sk_sendmsg_off) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_frag.offset; +} + +META_COLLECTOR(int_sk_write_pend) +{ + const struct sock *sk = skb_to_full_sk(skb); + + if (!sk) { + *err = -1; + return; + } + dst->value = sk->sk_write_pending; +} + +/************************************************************************** + * Meta value collectors assignment table + **************************************************************************/ + +struct meta_ops { + void (*get)(struct sk_buff *, struct tcf_pkt_info *, + struct meta_value *, struct meta_obj *, int *); +}; + +#define META_ID(name) TCF_META_ID_##name +#define META_FUNC(name) { .get = meta_##name } + +/* Meta value operations table listing all meta value collectors and + * assigns them to a type and meta id. */ +static struct meta_ops __meta_ops[TCF_META_TYPE_MAX + 1][TCF_META_ID_MAX + 1] = { + [TCF_META_TYPE_VAR] = { + [META_ID(DEV)] = META_FUNC(var_dev), + [META_ID(SK_BOUND_IF)] = META_FUNC(var_sk_bound_if), + }, + [TCF_META_TYPE_INT] = { + [META_ID(RANDOM)] = META_FUNC(int_random), + [META_ID(LOADAVG_0)] = META_FUNC(int_loadavg_0), + [META_ID(LOADAVG_1)] = META_FUNC(int_loadavg_1), + [META_ID(LOADAVG_2)] = META_FUNC(int_loadavg_2), + [META_ID(DEV)] = META_FUNC(int_dev), + [META_ID(PRIORITY)] = META_FUNC(int_priority), + [META_ID(PROTOCOL)] = META_FUNC(int_protocol), + [META_ID(PKTTYPE)] = META_FUNC(int_pkttype), + [META_ID(PKTLEN)] = META_FUNC(int_pktlen), + [META_ID(DATALEN)] = META_FUNC(int_datalen), + [META_ID(MACLEN)] = META_FUNC(int_maclen), + [META_ID(NFMARK)] = META_FUNC(int_mark), + [META_ID(TCINDEX)] = META_FUNC(int_tcindex), + [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), + [META_ID(RTIIF)] = META_FUNC(int_rtiif), + [META_ID(SK_FAMILY)] = META_FUNC(int_sk_family), + [META_ID(SK_STATE)] = META_FUNC(int_sk_state), + [META_ID(SK_REUSE)] = META_FUNC(int_sk_reuse), + [META_ID(SK_BOUND_IF)] = META_FUNC(int_sk_bound_if), + [META_ID(SK_REFCNT)] = META_FUNC(int_sk_refcnt), + [META_ID(SK_RCVBUF)] = META_FUNC(int_sk_rcvbuf), + [META_ID(SK_SNDBUF)] = META_FUNC(int_sk_sndbuf), + [META_ID(SK_SHUTDOWN)] = META_FUNC(int_sk_shutdown), + [META_ID(SK_PROTO)] = META_FUNC(int_sk_proto), + [META_ID(SK_TYPE)] = META_FUNC(int_sk_type), + [META_ID(SK_RMEM_ALLOC)] = META_FUNC(int_sk_rmem_alloc), + [META_ID(SK_WMEM_ALLOC)] = META_FUNC(int_sk_wmem_alloc), + [META_ID(SK_OMEM_ALLOC)] = META_FUNC(int_sk_omem_alloc), + [META_ID(SK_WMEM_QUEUED)] = META_FUNC(int_sk_wmem_queued), + [META_ID(SK_RCV_QLEN)] = META_FUNC(int_sk_rcv_qlen), + [META_ID(SK_SND_QLEN)] = META_FUNC(int_sk_snd_qlen), + [META_ID(SK_ERR_QLEN)] = META_FUNC(int_sk_err_qlen), + [META_ID(SK_FORWARD_ALLOCS)] = META_FUNC(int_sk_fwd_alloc), + [META_ID(SK_ALLOCS)] = META_FUNC(int_sk_alloc), + [META_ID(SK_HASH)] = META_FUNC(int_sk_hash), + [META_ID(SK_LINGERTIME)] = META_FUNC(int_sk_lingertime), + [META_ID(SK_ACK_BACKLOG)] = META_FUNC(int_sk_ack_bl), + [META_ID(SK_MAX_ACK_BACKLOG)] = META_FUNC(int_sk_max_ack_bl), + [META_ID(SK_PRIO)] = META_FUNC(int_sk_prio), + [META_ID(SK_RCVLOWAT)] = META_FUNC(int_sk_rcvlowat), + [META_ID(SK_RCVTIMEO)] = META_FUNC(int_sk_rcvtimeo), + [META_ID(SK_SNDTIMEO)] = META_FUNC(int_sk_sndtimeo), + [META_ID(SK_SENDMSG_OFF)] = META_FUNC(int_sk_sendmsg_off), + [META_ID(SK_WRITE_PENDING)] = META_FUNC(int_sk_write_pend), + [META_ID(VLAN_TAG)] = META_FUNC(int_vlan_tag), + [META_ID(RXHASH)] = META_FUNC(int_rxhash), + } +}; + +static inline struct meta_ops *meta_ops(struct meta_value *val) +{ + return &__meta_ops[meta_type(val)][meta_id(val)]; +} + +/************************************************************************** + * Type specific operations for TCF_META_TYPE_VAR + **************************************************************************/ + +static int meta_var_compare(struct meta_obj *a, struct meta_obj *b) +{ + int r = a->len - b->len; + + if (r == 0) + r = memcmp((void *) a->value, (void *) b->value, a->len); + + return r; +} + +static int meta_var_change(struct meta_value *dst, struct nlattr *nla) +{ + int len = nla_len(nla); + + dst->val = (unsigned long)kmemdup(nla_data(nla), len, GFP_KERNEL); + if (dst->val == 0UL) + return -ENOMEM; + dst->len = len; + return 0; +} + +static void meta_var_destroy(struct meta_value *v) +{ + kfree((void *) v->val); +} + +static void meta_var_apply_extras(struct meta_value *v, + struct meta_obj *dst) +{ + int shift = v->hdr.shift; + + if (shift && shift < dst->len) + dst->len -= shift; +} + +static int meta_var_dump(struct sk_buff *skb, struct meta_value *v, int tlv) +{ + if (v->val && v->len && + nla_put(skb, tlv, v->len, (void *) v->val)) + goto nla_put_failure; + return 0; + +nla_put_failure: + return -1; +} + +/************************************************************************** + * Type specific operations for TCF_META_TYPE_INT + **************************************************************************/ + +static int meta_int_compare(struct meta_obj *a, struct meta_obj *b) +{ + /* Let gcc optimize it, the unlikely is not really based on + * some numbers but jump free code for mismatches seems + * more logical. */ + if (unlikely(a->value == b->value)) + return 0; + else if (a->value < b->value) + return -1; + else + return 1; +} + +static int meta_int_change(struct meta_value *dst, struct nlattr *nla) +{ + if (nla_len(nla) >= sizeof(unsigned long)) { + dst->val = *(unsigned long *) nla_data(nla); + dst->len = sizeof(unsigned long); + } else if (nla_len(nla) == sizeof(u32)) { + dst->val = nla_get_u32(nla); + dst->len = sizeof(u32); + } else + return -EINVAL; + + return 0; +} + +static void meta_int_apply_extras(struct meta_value *v, + struct meta_obj *dst) +{ + if (v->hdr.shift) + dst->value >>= v->hdr.shift; + + if (v->val) + dst->value &= v->val; +} + +static int meta_int_dump(struct sk_buff *skb, struct meta_value *v, int tlv) +{ + if (v->len == sizeof(unsigned long)) { + if (nla_put(skb, tlv, sizeof(unsigned long), &v->val)) + goto nla_put_failure; + } else if (v->len == sizeof(u32)) { + if (nla_put_u32(skb, tlv, v->val)) + goto nla_put_failure; + } + + return 0; + +nla_put_failure: + return -1; +} + +/************************************************************************** + * Type specific operations table + **************************************************************************/ + +struct meta_type_ops { + void (*destroy)(struct meta_value *); + int (*compare)(struct meta_obj *, struct meta_obj *); + int (*change)(struct meta_value *, struct nlattr *); + void (*apply_extras)(struct meta_value *, struct meta_obj *); + int (*dump)(struct sk_buff *, struct meta_value *, int); +}; + +static const struct meta_type_ops __meta_type_ops[TCF_META_TYPE_MAX + 1] = { + [TCF_META_TYPE_VAR] = { + .destroy = meta_var_destroy, + .compare = meta_var_compare, + .change = meta_var_change, + .apply_extras = meta_var_apply_extras, + .dump = meta_var_dump + }, + [TCF_META_TYPE_INT] = { + .compare = meta_int_compare, + .change = meta_int_change, + .apply_extras = meta_int_apply_extras, + .dump = meta_int_dump + } +}; + +static inline const struct meta_type_ops *meta_type_ops(struct meta_value *v) +{ + return &__meta_type_ops[meta_type(v)]; +} + +/************************************************************************** + * Core + **************************************************************************/ + +static int meta_get(struct sk_buff *skb, struct tcf_pkt_info *info, + struct meta_value *v, struct meta_obj *dst) +{ + int err = 0; + + if (meta_id(v) == TCF_META_ID_VALUE) { + dst->value = v->val; + dst->len = v->len; + return 0; + } + + meta_ops(v)->get(skb, info, v, dst, &err); + if (err < 0) + return err; + + if (meta_type_ops(v)->apply_extras) + meta_type_ops(v)->apply_extras(v, dst); + + return 0; +} + +static int em_meta_match(struct sk_buff *skb, struct tcf_ematch *m, + struct tcf_pkt_info *info) +{ + int r; + struct meta_match *meta = (struct meta_match *) m->data; + struct meta_obj l_value, r_value; + + if (meta_get(skb, info, &meta->lvalue, &l_value) < 0 || + meta_get(skb, info, &meta->rvalue, &r_value) < 0) + return 0; + + r = meta_type_ops(&meta->lvalue)->compare(&l_value, &r_value); + + switch (meta->lvalue.hdr.op) { + case TCF_EM_OPND_EQ: + return !r; + case TCF_EM_OPND_LT: + return r < 0; + case TCF_EM_OPND_GT: + return r > 0; + } + + return 0; +} + +static void meta_delete(struct meta_match *meta) +{ + if (meta) { + const struct meta_type_ops *ops = meta_type_ops(&meta->lvalue); + + if (ops && ops->destroy) { + ops->destroy(&meta->lvalue); + ops->destroy(&meta->rvalue); + } + } + + kfree(meta); +} + +static inline int meta_change_data(struct meta_value *dst, struct nlattr *nla) +{ + if (nla) { + if (nla_len(nla) == 0) + return -EINVAL; + + return meta_type_ops(dst)->change(dst, nla); + } + + return 0; +} + +static inline int meta_is_supported(struct meta_value *val) +{ + return !meta_id(val) || meta_ops(val)->get; +} + +static const struct nla_policy meta_policy[TCA_EM_META_MAX + 1] = { + [TCA_EM_META_HDR] = { .len = sizeof(struct tcf_meta_hdr) }, +}; + +static int em_meta_change(struct net *net, void *data, int len, + struct tcf_ematch *m) +{ + int err; + struct nlattr *tb[TCA_EM_META_MAX + 1]; + struct tcf_meta_hdr *hdr; + struct meta_match *meta = NULL; + + err = nla_parse(tb, TCA_EM_META_MAX, data, len, meta_policy, NULL); + if (err < 0) + goto errout; + + err = -EINVAL; + if (tb[TCA_EM_META_HDR] == NULL) + goto errout; + hdr = nla_data(tb[TCA_EM_META_HDR]); + + if (TCF_META_TYPE(hdr->left.kind) != TCF_META_TYPE(hdr->right.kind) || + TCF_META_TYPE(hdr->left.kind) > TCF_META_TYPE_MAX || + TCF_META_ID(hdr->left.kind) > TCF_META_ID_MAX || + TCF_META_ID(hdr->right.kind) > TCF_META_ID_MAX) + goto errout; + + meta = kzalloc(sizeof(*meta), GFP_KERNEL); + if (meta == NULL) { + err = -ENOMEM; + goto errout; + } + + memcpy(&meta->lvalue.hdr, &hdr->left, sizeof(hdr->left)); + memcpy(&meta->rvalue.hdr, &hdr->right, sizeof(hdr->right)); + + if (!meta_is_supported(&meta->lvalue) || + !meta_is_supported(&meta->rvalue)) { + err = -EOPNOTSUPP; + goto errout; + } + + if (meta_change_data(&meta->lvalue, tb[TCA_EM_META_LVALUE]) < 0 || + meta_change_data(&meta->rvalue, tb[TCA_EM_META_RVALUE]) < 0) + goto errout; + + m->datalen = sizeof(*meta); + m->data = (unsigned long) meta; + + err = 0; +errout: + if (err && meta) + meta_delete(meta); + return err; +} + +static void em_meta_destroy(struct tcf_ematch *m) +{ + if (m) + meta_delete((struct meta_match *) m->data); +} + +static int em_meta_dump(struct sk_buff *skb, struct tcf_ematch *em) +{ + struct meta_match *meta = (struct meta_match *) em->data; + struct tcf_meta_hdr hdr; + const struct meta_type_ops *ops; + + memset(&hdr, 0, sizeof(hdr)); + memcpy(&hdr.left, &meta->lvalue.hdr, sizeof(hdr.left)); + memcpy(&hdr.right, &meta->rvalue.hdr, sizeof(hdr.right)); + + if (nla_put(skb, TCA_EM_META_HDR, sizeof(hdr), &hdr)) + goto nla_put_failure; + + ops = meta_type_ops(&meta->lvalue); + if (ops->dump(skb, &meta->lvalue, TCA_EM_META_LVALUE) < 0 || + ops->dump(skb, &meta->rvalue, TCA_EM_META_RVALUE) < 0) + goto nla_put_failure; + + return 0; + +nla_put_failure: + return -1; +} + +static struct tcf_ematch_ops em_meta_ops = { + .kind = TCF_EM_META, + .change = em_meta_change, + .match = em_meta_match, + .destroy = em_meta_destroy, + .dump = em_meta_dump, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_meta_ops.link) +}; + +static int __init init_em_meta(void) +{ + return tcf_em_register(&em_meta_ops); +} + +static void __exit exit_em_meta(void) +{ + tcf_em_unregister(&em_meta_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_meta); +module_exit(exit_em_meta); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_META); diff --git a/net/sched/em_nbyte.c b/net/sched/em_nbyte.c new file mode 100644 index 000000000..07c10bac0 --- /dev/null +++ b/net/sched/em_nbyte.c @@ -0,0 +1,80 @@ +/* + * net/sched/em_nbyte.c N-Byte ematch + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/gfp.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <linux/tc_ematch/tc_em_nbyte.h> +#include <net/pkt_cls.h> + +struct nbyte_data { + struct tcf_em_nbyte hdr; + char pattern[0]; +}; + +static int em_nbyte_change(struct net *net, void *data, int data_len, + struct tcf_ematch *em) +{ + struct tcf_em_nbyte *nbyte = data; + + if (data_len < sizeof(*nbyte) || + data_len < (sizeof(*nbyte) + nbyte->len)) + return -EINVAL; + + em->datalen = sizeof(*nbyte) + nbyte->len; + em->data = (unsigned long)kmemdup(data, em->datalen, GFP_KERNEL); + if (em->data == 0UL) + return -ENOBUFS; + + return 0; +} + +static int em_nbyte_match(struct sk_buff *skb, struct tcf_ematch *em, + struct tcf_pkt_info *info) +{ + struct nbyte_data *nbyte = (struct nbyte_data *) em->data; + unsigned char *ptr = tcf_get_base_ptr(skb, nbyte->hdr.layer); + + ptr += nbyte->hdr.off; + + if (!tcf_valid_offset(skb, ptr, nbyte->hdr.len)) + return 0; + + return !memcmp(ptr, nbyte->pattern, nbyte->hdr.len); +} + +static struct tcf_ematch_ops em_nbyte_ops = { + .kind = TCF_EM_NBYTE, + .change = em_nbyte_change, + .match = em_nbyte_match, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_nbyte_ops.link) +}; + +static int __init init_em_nbyte(void) +{ + return tcf_em_register(&em_nbyte_ops); +} + +static void __exit exit_em_nbyte(void) +{ + tcf_em_unregister(&em_nbyte_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_nbyte); +module_exit(exit_em_nbyte); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_NBYTE); diff --git a/net/sched/em_text.c b/net/sched/em_text.c new file mode 100644 index 000000000..73e2ed576 --- /dev/null +++ b/net/sched/em_text.c @@ -0,0 +1,157 @@ +/* + * net/sched/em_text.c Textsearch ematch + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <linux/textsearch.h> +#include <linux/tc_ematch/tc_em_text.h> +#include <net/pkt_cls.h> + +struct text_match { + u16 from_offset; + u16 to_offset; + u8 from_layer; + u8 to_layer; + struct ts_config *config; +}; + +#define EM_TEXT_PRIV(m) ((struct text_match *) (m)->data) + +static int em_text_match(struct sk_buff *skb, struct tcf_ematch *m, + struct tcf_pkt_info *info) +{ + struct text_match *tm = EM_TEXT_PRIV(m); + int from, to; + + from = tcf_get_base_ptr(skb, tm->from_layer) - skb->data; + from += tm->from_offset; + + to = tcf_get_base_ptr(skb, tm->to_layer) - skb->data; + to += tm->to_offset; + + return skb_find_text(skb, from, to, tm->config) != UINT_MAX; +} + +static int em_text_change(struct net *net, void *data, int len, + struct tcf_ematch *m) +{ + struct text_match *tm; + struct tcf_em_text *conf = data; + struct ts_config *ts_conf; + int flags = 0; + + if (len < sizeof(*conf) || len < (sizeof(*conf) + conf->pattern_len)) + return -EINVAL; + + if (conf->from_layer > conf->to_layer) + return -EINVAL; + + if (conf->from_layer == conf->to_layer && + conf->from_offset > conf->to_offset) + return -EINVAL; + +retry: + ts_conf = textsearch_prepare(conf->algo, (u8 *) conf + sizeof(*conf), + conf->pattern_len, GFP_KERNEL, flags); + + if (flags & TS_AUTOLOAD) + rtnl_lock(); + + if (IS_ERR(ts_conf)) { + if (PTR_ERR(ts_conf) == -ENOENT && !(flags & TS_AUTOLOAD)) { + rtnl_unlock(); + flags |= TS_AUTOLOAD; + goto retry; + } else + return PTR_ERR(ts_conf); + } else if (flags & TS_AUTOLOAD) { + textsearch_destroy(ts_conf); + return -EAGAIN; + } + + tm = kmalloc(sizeof(*tm), GFP_KERNEL); + if (tm == NULL) { + textsearch_destroy(ts_conf); + return -ENOBUFS; + } + + tm->from_offset = conf->from_offset; + tm->to_offset = conf->to_offset; + tm->from_layer = conf->from_layer; + tm->to_layer = conf->to_layer; + tm->config = ts_conf; + + m->datalen = sizeof(*tm); + m->data = (unsigned long) tm; + + return 0; +} + +static void em_text_destroy(struct tcf_ematch *m) +{ + if (EM_TEXT_PRIV(m) && EM_TEXT_PRIV(m)->config) + textsearch_destroy(EM_TEXT_PRIV(m)->config); +} + +static int em_text_dump(struct sk_buff *skb, struct tcf_ematch *m) +{ + struct text_match *tm = EM_TEXT_PRIV(m); + struct tcf_em_text conf; + + strncpy(conf.algo, tm->config->ops->name, sizeof(conf.algo) - 1); + conf.from_offset = tm->from_offset; + conf.to_offset = tm->to_offset; + conf.from_layer = tm->from_layer; + conf.to_layer = tm->to_layer; + conf.pattern_len = textsearch_get_pattern_len(tm->config); + conf.pad = 0; + + if (nla_put_nohdr(skb, sizeof(conf), &conf) < 0) + goto nla_put_failure; + if (nla_append(skb, conf.pattern_len, + textsearch_get_pattern(tm->config)) < 0) + goto nla_put_failure; + return 0; + +nla_put_failure: + return -1; +} + +static struct tcf_ematch_ops em_text_ops = { + .kind = TCF_EM_TEXT, + .change = em_text_change, + .match = em_text_match, + .destroy = em_text_destroy, + .dump = em_text_dump, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_text_ops.link) +}; + +static int __init init_em_text(void) +{ + return tcf_em_register(&em_text_ops); +} + +static void __exit exit_em_text(void) +{ + tcf_em_unregister(&em_text_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_text); +module_exit(exit_em_text); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_TEXT); diff --git a/net/sched/em_u32.c b/net/sched/em_u32.c new file mode 100644 index 000000000..797bdb88c --- /dev/null +++ b/net/sched/em_u32.c @@ -0,0 +1,64 @@ +/* + * net/sched/em_u32.c U32 Ematch + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + * Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + * Based on net/sched/cls_u32.c + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <net/pkt_cls.h> + +static int em_u32_match(struct sk_buff *skb, struct tcf_ematch *em, + struct tcf_pkt_info *info) +{ + struct tc_u32_key *key = (struct tc_u32_key *) em->data; + const unsigned char *ptr = skb_network_header(skb); + + if (info) { + if (info->ptr) + ptr = info->ptr; + ptr += (info->nexthdr & key->offmask); + } + + ptr += key->off; + + if (!tcf_valid_offset(skb, ptr, sizeof(u32))) + return 0; + + return !(((*(__be32 *) ptr) ^ key->val) & key->mask); +} + +static struct tcf_ematch_ops em_u32_ops = { + .kind = TCF_EM_U32, + .datalen = sizeof(struct tc_u32_key), + .match = em_u32_match, + .owner = THIS_MODULE, + .link = LIST_HEAD_INIT(em_u32_ops.link) +}; + +static int __init init_em_u32(void) +{ + return tcf_em_register(&em_u32_ops); +} + +static void __exit exit_em_u32(void) +{ + tcf_em_unregister(&em_u32_ops); +} + +MODULE_LICENSE("GPL"); + +module_init(init_em_u32); +module_exit(exit_em_u32); + +MODULE_ALIAS_TCF_EMATCH(TCF_EM_U32); diff --git a/net/sched/ematch.c b/net/sched/ematch.c new file mode 100644 index 000000000..113a133ee --- /dev/null +++ b/net/sched/ematch.c @@ -0,0 +1,552 @@ +/* + * net/sched/ematch.c Extended Match API + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + * + * ========================================================================== + * + * An extended match (ematch) is a small classification tool not worth + * writing a full classifier for. Ematches can be interconnected to form + * a logic expression and get attached to classifiers to extend their + * functionatlity. + * + * The userspace part transforms the logic expressions into an array + * consisting of multiple sequences of interconnected ematches separated + * by markers. Precedence is implemented by a special ematch kind + * referencing a sequence beyond the marker of the current sequence + * causing the current position in the sequence to be pushed onto a stack + * to allow the current position to be overwritten by the position referenced + * in the special ematch. Matching continues in the new sequence until a + * marker is reached causing the position to be restored from the stack. + * + * Example: + * A AND (B1 OR B2) AND C AND D + * + * ------->-PUSH------- + * -->-- / -->-- \ -->-- + * / \ / / \ \ / \ + * +-------+-------+-------+-------+-------+--------+ + * | A AND | B AND | C AND | D END | B1 OR | B2 END | + * +-------+-------+-------+-------+-------+--------+ + * \ / + * --------<-POP--------- + * + * where B is a virtual ematch referencing to sequence starting with B1. + * + * ========================================================================== + * + * How to write an ematch in 60 seconds + * ------------------------------------ + * + * 1) Provide a matcher function: + * static int my_match(struct sk_buff *skb, struct tcf_ematch *m, + * struct tcf_pkt_info *info) + * { + * struct mydata *d = (struct mydata *) m->data; + * + * if (...matching goes here...) + * return 1; + * else + * return 0; + * } + * + * 2) Fill out a struct tcf_ematch_ops: + * static struct tcf_ematch_ops my_ops = { + * .kind = unique id, + * .datalen = sizeof(struct mydata), + * .match = my_match, + * .owner = THIS_MODULE, + * }; + * + * 3) Register/Unregister your ematch: + * static int __init init_my_ematch(void) + * { + * return tcf_em_register(&my_ops); + * } + * + * static void __exit exit_my_ematch(void) + * { + * tcf_em_unregister(&my_ops); + * } + * + * module_init(init_my_ematch); + * module_exit(exit_my_ematch); + * + * 4) By now you should have two more seconds left, barely enough to + * open up a beer to watch the compilation going. + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/rtnetlink.h> +#include <linux/skbuff.h> +#include <net/pkt_cls.h> + +static LIST_HEAD(ematch_ops); +static DEFINE_RWLOCK(ematch_mod_lock); + +static struct tcf_ematch_ops *tcf_em_lookup(u16 kind) +{ + struct tcf_ematch_ops *e = NULL; + + read_lock(&ematch_mod_lock); + list_for_each_entry(e, &ematch_ops, link) { + if (kind == e->kind) { + if (!try_module_get(e->owner)) + e = NULL; + read_unlock(&ematch_mod_lock); + return e; + } + } + read_unlock(&ematch_mod_lock); + + return NULL; +} + +/** + * tcf_em_register - register an extended match + * + * @ops: ematch operations lookup table + * + * This function must be called by ematches to announce their presence. + * The given @ops must have kind set to a unique identifier and the + * callback match() must be implemented. All other callbacks are optional + * and a fallback implementation is used instead. + * + * Returns -EEXISTS if an ematch of the same kind has already registered. + */ +int tcf_em_register(struct tcf_ematch_ops *ops) +{ + int err = -EEXIST; + struct tcf_ematch_ops *e; + + if (ops->match == NULL) + return -EINVAL; + + write_lock(&ematch_mod_lock); + list_for_each_entry(e, &ematch_ops, link) + if (ops->kind == e->kind) + goto errout; + + list_add_tail(&ops->link, &ematch_ops); + err = 0; +errout: + write_unlock(&ematch_mod_lock); + return err; +} +EXPORT_SYMBOL(tcf_em_register); + +/** + * tcf_em_unregister - unregster and extended match + * + * @ops: ematch operations lookup table + * + * This function must be called by ematches to announce their disappearance + * for examples when the module gets unloaded. The @ops parameter must be + * the same as the one used for registration. + * + * Returns -ENOENT if no matching ematch was found. + */ +void tcf_em_unregister(struct tcf_ematch_ops *ops) +{ + write_lock(&ematch_mod_lock); + list_del(&ops->link); + write_unlock(&ematch_mod_lock); +} +EXPORT_SYMBOL(tcf_em_unregister); + +static inline struct tcf_ematch *tcf_em_get_match(struct tcf_ematch_tree *tree, + int index) +{ + return &tree->matches[index]; +} + + +static int tcf_em_validate(struct tcf_proto *tp, + struct tcf_ematch_tree_hdr *tree_hdr, + struct tcf_ematch *em, struct nlattr *nla, int idx) +{ + int err = -EINVAL; + struct tcf_ematch_hdr *em_hdr = nla_data(nla); + int data_len = nla_len(nla) - sizeof(*em_hdr); + void *data = (void *) em_hdr + sizeof(*em_hdr); + struct net *net = tp->chain->block->net; + + if (!TCF_EM_REL_VALID(em_hdr->flags)) + goto errout; + + if (em_hdr->kind == TCF_EM_CONTAINER) { + /* Special ematch called "container", carries an index + * referencing an external ematch sequence. + */ + u32 ref; + + if (data_len < sizeof(ref)) + goto errout; + ref = *(u32 *) data; + + if (ref >= tree_hdr->nmatches) + goto errout; + + /* We do not allow backward jumps to avoid loops and jumps + * to our own position are of course illegal. + */ + if (ref <= idx) + goto errout; + + + em->data = ref; + } else { + /* Note: This lookup will increase the module refcnt + * of the ematch module referenced. In case of a failure, + * a destroy function is called by the underlying layer + * which automatically releases the reference again, therefore + * the module MUST not be given back under any circumstances + * here. Be aware, the destroy function assumes that the + * module is held if the ops field is non zero. + */ + em->ops = tcf_em_lookup(em_hdr->kind); + + if (em->ops == NULL) { + err = -ENOENT; +#ifdef CONFIG_MODULES + __rtnl_unlock(); + request_module("ematch-kind-%u", em_hdr->kind); + rtnl_lock(); + em->ops = tcf_em_lookup(em_hdr->kind); + if (em->ops) { + /* We dropped the RTNL mutex in order to + * perform the module load. Tell the caller + * to replay the request. + */ + module_put(em->ops->owner); + em->ops = NULL; + err = -EAGAIN; + } +#endif + goto errout; + } + + /* ematch module provides expected length of data, so we + * can do a basic sanity check. + */ + if (em->ops->datalen && data_len < em->ops->datalen) + goto errout; + + if (em->ops->change) { + err = -EINVAL; + if (em_hdr->flags & TCF_EM_SIMPLE) + goto errout; + err = em->ops->change(net, data, data_len, em); + if (err < 0) + goto errout; + } else if (data_len > 0) { + /* ematch module doesn't provide an own change + * procedure and expects us to allocate and copy + * the ematch data. + * + * TCF_EM_SIMPLE may be specified stating that the + * data only consists of a u32 integer and the module + * does not expected a memory reference but rather + * the value carried. + */ + if (em_hdr->flags & TCF_EM_SIMPLE) { + if (data_len < sizeof(u32)) + goto errout; + em->data = *(u32 *) data; + } else { + void *v = kmemdup(data, data_len, GFP_KERNEL); + if (v == NULL) { + err = -ENOBUFS; + goto errout; + } + em->data = (unsigned long) v; + } + em->datalen = data_len; + } + } + + em->matchid = em_hdr->matchid; + em->flags = em_hdr->flags; + em->net = net; + + err = 0; +errout: + return err; +} + +static const struct nla_policy em_policy[TCA_EMATCH_TREE_MAX + 1] = { + [TCA_EMATCH_TREE_HDR] = { .len = sizeof(struct tcf_ematch_tree_hdr) }, + [TCA_EMATCH_TREE_LIST] = { .type = NLA_NESTED }, +}; + +/** + * tcf_em_tree_validate - validate ematch config TLV and build ematch tree + * + * @tp: classifier kind handle + * @nla: ematch tree configuration TLV + * @tree: destination ematch tree variable to store the resulting + * ematch tree. + * + * This function validates the given configuration TLV @nla and builds an + * ematch tree in @tree. The resulting tree must later be copied into + * the private classifier data using tcf_em_tree_change(). You MUST NOT + * provide the ematch tree variable of the private classifier data directly, + * the changes would not be locked properly. + * + * Returns a negative error code if the configuration TLV contains errors. + */ +int tcf_em_tree_validate(struct tcf_proto *tp, struct nlattr *nla, + struct tcf_ematch_tree *tree) +{ + int idx, list_len, matches_len, err; + struct nlattr *tb[TCA_EMATCH_TREE_MAX + 1]; + struct nlattr *rt_match, *rt_hdr, *rt_list; + struct tcf_ematch_tree_hdr *tree_hdr; + struct tcf_ematch *em; + + memset(tree, 0, sizeof(*tree)); + if (!nla) + return 0; + + err = nla_parse_nested(tb, TCA_EMATCH_TREE_MAX, nla, em_policy, NULL); + if (err < 0) + goto errout; + + err = -EINVAL; + rt_hdr = tb[TCA_EMATCH_TREE_HDR]; + rt_list = tb[TCA_EMATCH_TREE_LIST]; + + if (rt_hdr == NULL || rt_list == NULL) + goto errout; + + tree_hdr = nla_data(rt_hdr); + memcpy(&tree->hdr, tree_hdr, sizeof(*tree_hdr)); + + rt_match = nla_data(rt_list); + list_len = nla_len(rt_list); + matches_len = tree_hdr->nmatches * sizeof(*em); + + tree->matches = kzalloc(matches_len, GFP_KERNEL); + if (tree->matches == NULL) + goto errout; + + /* We do not use nla_parse_nested here because the maximum + * number of attributes is unknown. This saves us the allocation + * for a tb buffer which would serve no purpose at all. + * + * The array of rt attributes is parsed in the order as they are + * provided, their type must be incremental from 1 to n. Even + * if it does not serve any real purpose, a failure of sticking + * to this policy will result in parsing failure. + */ + for (idx = 0; nla_ok(rt_match, list_len); idx++) { + err = -EINVAL; + + if (rt_match->nla_type != (idx + 1)) + goto errout_abort; + + if (idx >= tree_hdr->nmatches) + goto errout_abort; + + if (nla_len(rt_match) < sizeof(struct tcf_ematch_hdr)) + goto errout_abort; + + em = tcf_em_get_match(tree, idx); + + err = tcf_em_validate(tp, tree_hdr, em, rt_match, idx); + if (err < 0) + goto errout_abort; + + rt_match = nla_next(rt_match, &list_len); + } + + /* Check if the number of matches provided by userspace actually + * complies with the array of matches. The number was used for + * the validation of references and a mismatch could lead to + * undefined references during the matching process. + */ + if (idx != tree_hdr->nmatches) { + err = -EINVAL; + goto errout_abort; + } + + err = 0; +errout: + return err; + +errout_abort: + tcf_em_tree_destroy(tree); + return err; +} +EXPORT_SYMBOL(tcf_em_tree_validate); + +/** + * tcf_em_tree_destroy - destroy an ematch tree + * + * @tp: classifier kind handle + * @tree: ematch tree to be deleted + * + * This functions destroys an ematch tree previously created by + * tcf_em_tree_validate()/tcf_em_tree_change(). You must ensure that + * the ematch tree is not in use before calling this function. + */ +void tcf_em_tree_destroy(struct tcf_ematch_tree *tree) +{ + int i; + + if (tree->matches == NULL) + return; + + for (i = 0; i < tree->hdr.nmatches; i++) { + struct tcf_ematch *em = tcf_em_get_match(tree, i); + + if (em->ops) { + if (em->ops->destroy) + em->ops->destroy(em); + else if (!tcf_em_is_simple(em)) + kfree((void *) em->data); + module_put(em->ops->owner); + } + } + + tree->hdr.nmatches = 0; + kfree(tree->matches); + tree->matches = NULL; +} +EXPORT_SYMBOL(tcf_em_tree_destroy); + +/** + * tcf_em_tree_dump - dump ematch tree into a rtnl message + * + * @skb: skb holding the rtnl message + * @t: ematch tree to be dumped + * @tlv: TLV type to be used to encapsulate the tree + * + * This function dumps a ematch tree into a rtnl message. It is valid to + * call this function while the ematch tree is in use. + * + * Returns -1 if the skb tailroom is insufficient. + */ +int tcf_em_tree_dump(struct sk_buff *skb, struct tcf_ematch_tree *tree, int tlv) +{ + int i; + u8 *tail; + struct nlattr *top_start; + struct nlattr *list_start; + + top_start = nla_nest_start(skb, tlv); + if (top_start == NULL) + goto nla_put_failure; + + if (nla_put(skb, TCA_EMATCH_TREE_HDR, sizeof(tree->hdr), &tree->hdr)) + goto nla_put_failure; + + list_start = nla_nest_start(skb, TCA_EMATCH_TREE_LIST); + if (list_start == NULL) + goto nla_put_failure; + + tail = skb_tail_pointer(skb); + for (i = 0; i < tree->hdr.nmatches; i++) { + struct nlattr *match_start = (struct nlattr *)tail; + struct tcf_ematch *em = tcf_em_get_match(tree, i); + struct tcf_ematch_hdr em_hdr = { + .kind = em->ops ? em->ops->kind : TCF_EM_CONTAINER, + .matchid = em->matchid, + .flags = em->flags + }; + + if (nla_put(skb, i + 1, sizeof(em_hdr), &em_hdr)) + goto nla_put_failure; + + if (em->ops && em->ops->dump) { + if (em->ops->dump(skb, em) < 0) + goto nla_put_failure; + } else if (tcf_em_is_container(em) || tcf_em_is_simple(em)) { + u32 u = em->data; + nla_put_nohdr(skb, sizeof(u), &u); + } else if (em->datalen > 0) + nla_put_nohdr(skb, em->datalen, (void *) em->data); + + tail = skb_tail_pointer(skb); + match_start->nla_len = tail - (u8 *)match_start; + } + + nla_nest_end(skb, list_start); + nla_nest_end(skb, top_start); + + return 0; + +nla_put_failure: + return -1; +} +EXPORT_SYMBOL(tcf_em_tree_dump); + +static inline int tcf_em_match(struct sk_buff *skb, struct tcf_ematch *em, + struct tcf_pkt_info *info) +{ + int r = em->ops->match(skb, em, info); + + return tcf_em_is_inverted(em) ? !r : r; +} + +/* Do not use this function directly, use tcf_em_tree_match instead */ +int __tcf_em_tree_match(struct sk_buff *skb, struct tcf_ematch_tree *tree, + struct tcf_pkt_info *info) +{ + int stackp = 0, match_idx = 0, res = 0; + struct tcf_ematch *cur_match; + int stack[CONFIG_NET_EMATCH_STACK]; + +proceed: + while (match_idx < tree->hdr.nmatches) { + cur_match = tcf_em_get_match(tree, match_idx); + + if (tcf_em_is_container(cur_match)) { + if (unlikely(stackp >= CONFIG_NET_EMATCH_STACK)) + goto stack_overflow; + + stack[stackp++] = match_idx; + match_idx = cur_match->data; + goto proceed; + } + + res = tcf_em_match(skb, cur_match, info); + + if (tcf_em_early_end(cur_match, res)) + break; + + match_idx++; + } + +pop_stack: + if (stackp > 0) { + match_idx = stack[--stackp]; + cur_match = tcf_em_get_match(tree, match_idx); + + if (tcf_em_is_inverted(cur_match)) + res = !res; + + if (tcf_em_early_end(cur_match, res)) { + goto pop_stack; + } else { + match_idx++; + goto proceed; + } + } + + return res; + +stack_overflow: + net_warn_ratelimited("tc ematch: local stack overflow, increase NET_EMATCH_STACK\n"); + return -1; +} +EXPORT_SYMBOL(__tcf_em_tree_match); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c new file mode 100644 index 000000000..424e70907 --- /dev/null +++ b/net/sched/sch_api.c @@ -0,0 +1,2209 @@ +/* + * net/sched/sch_api.c Packet scheduler API. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + * Fixes: + * + * Rani Assaf <rani@magic.metawire.com> :980802: JIFFIES and CPU clock sources are repaired. + * Eduardo J. Blanco <ejbs@netlabs.com.uy> :990222: kmod support + * Jamal Hadi Salim <hadi@nortelnetworks.com>: 990601: ingress support + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/init.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/kmod.h> +#include <linux/list.h> +#include <linux/hrtimer.h> +#include <linux/lockdep.h> +#include <linux/slab.h> +#include <linux/hashtable.h> + +#include <net/net_namespace.h> +#include <net/sock.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +/* + + Short review. + ------------- + + This file consists of two interrelated parts: + + 1. queueing disciplines manager frontend. + 2. traffic classes manager frontend. + + Generally, queueing discipline ("qdisc") is a black box, + which is able to enqueue packets and to dequeue them (when + device is ready to send something) in order and at times + determined by algorithm hidden in it. + + qdisc's are divided to two categories: + - "queues", which have no internal structure visible from outside. + - "schedulers", which split all the packets to "traffic classes", + using "packet classifiers" (look at cls_api.c) + + In turn, classes may have child qdiscs (as rule, queues) + attached to them etc. etc. etc. + + The goal of the routines in this file is to translate + information supplied by user in the form of handles + to more intelligible for kernel form, to make some sanity + checks and part of work, which is common to all qdiscs + and to provide rtnetlink notifications. + + All real intelligent work is done inside qdisc modules. + + + + Every discipline has two major routines: enqueue and dequeue. + + ---dequeue + + dequeue usually returns a skb to send. It is allowed to return NULL, + but it does not mean that queue is empty, it just means that + discipline does not want to send anything this time. + Queue is really empty if q->q.qlen == 0. + For complicated disciplines with multiple queues q->q is not + real packet queue, but however q->q.qlen must be valid. + + ---enqueue + + enqueue returns 0, if packet was enqueued successfully. + If packet (this one or another one) was dropped, it returns + not zero error code. + NET_XMIT_DROP - this packet dropped + Expected action: do not backoff, but wait until queue will clear. + NET_XMIT_CN - probably this packet enqueued, but another one dropped. + Expected action: backoff or ignore + + Auxiliary routines: + + ---peek + + like dequeue but without removing a packet from the queue + + ---reset + + returns qdisc to initial state: purge all buffers, clear all + timers, counters (except for statistics) etc. + + ---init + + initializes newly created qdisc. + + ---destroy + + destroys resources allocated by init and during lifetime of qdisc. + + ---change + + changes qdisc parameters. + */ + +/* Protects list of registered TC modules. It is pure SMP lock. */ +static DEFINE_RWLOCK(qdisc_mod_lock); + + +/************************************************ + * Queueing disciplines manipulation. * + ************************************************/ + + +/* The list of all installed queueing disciplines. */ + +static struct Qdisc_ops *qdisc_base; + +/* Register/unregister queueing discipline */ + +int register_qdisc(struct Qdisc_ops *qops) +{ + struct Qdisc_ops *q, **qp; + int rc = -EEXIST; + + write_lock(&qdisc_mod_lock); + for (qp = &qdisc_base; (q = *qp) != NULL; qp = &q->next) + if (!strcmp(qops->id, q->id)) + goto out; + + if (qops->enqueue == NULL) + qops->enqueue = noop_qdisc_ops.enqueue; + if (qops->peek == NULL) { + if (qops->dequeue == NULL) + qops->peek = noop_qdisc_ops.peek; + else + goto out_einval; + } + if (qops->dequeue == NULL) + qops->dequeue = noop_qdisc_ops.dequeue; + + if (qops->cl_ops) { + const struct Qdisc_class_ops *cops = qops->cl_ops; + + if (!(cops->find && cops->walk && cops->leaf)) + goto out_einval; + + if (cops->tcf_block && !(cops->bind_tcf && cops->unbind_tcf)) + goto out_einval; + } + + qops->next = NULL; + *qp = qops; + rc = 0; +out: + write_unlock(&qdisc_mod_lock); + return rc; + +out_einval: + rc = -EINVAL; + goto out; +} +EXPORT_SYMBOL(register_qdisc); + +int unregister_qdisc(struct Qdisc_ops *qops) +{ + struct Qdisc_ops *q, **qp; + int err = -ENOENT; + + write_lock(&qdisc_mod_lock); + for (qp = &qdisc_base; (q = *qp) != NULL; qp = &q->next) + if (q == qops) + break; + if (q) { + *qp = q->next; + q->next = NULL; + err = 0; + } + write_unlock(&qdisc_mod_lock); + return err; +} +EXPORT_SYMBOL(unregister_qdisc); + +/* Get default qdisc if not otherwise specified */ +void qdisc_get_default(char *name, size_t len) +{ + read_lock(&qdisc_mod_lock); + strlcpy(name, default_qdisc_ops->id, len); + read_unlock(&qdisc_mod_lock); +} + +static struct Qdisc_ops *qdisc_lookup_default(const char *name) +{ + struct Qdisc_ops *q = NULL; + + for (q = qdisc_base; q; q = q->next) { + if (!strcmp(name, q->id)) { + if (!try_module_get(q->owner)) + q = NULL; + break; + } + } + + return q; +} + +/* Set new default qdisc to use */ +int qdisc_set_default(const char *name) +{ + const struct Qdisc_ops *ops; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + write_lock(&qdisc_mod_lock); + ops = qdisc_lookup_default(name); + if (!ops) { + /* Not found, drop lock and try to load module */ + write_unlock(&qdisc_mod_lock); + request_module("sch_%s", name); + write_lock(&qdisc_mod_lock); + + ops = qdisc_lookup_default(name); + } + + if (ops) { + /* Set new default */ + module_put(default_qdisc_ops->owner); + default_qdisc_ops = ops; + } + write_unlock(&qdisc_mod_lock); + + return ops ? 0 : -ENOENT; +} + +#ifdef CONFIG_NET_SCH_DEFAULT +/* Set default value from kernel config */ +static int __init sch_default_qdisc(void) +{ + return qdisc_set_default(CONFIG_DEFAULT_NET_SCH); +} +late_initcall(sch_default_qdisc); +#endif + +/* We know handle. Find qdisc among all qdisc's attached to device + * (root qdisc, all its children, children of children etc.) + * Note: caller either uses rtnl or rcu_read_lock() + */ + +static struct Qdisc *qdisc_match_from_root(struct Qdisc *root, u32 handle) +{ + struct Qdisc *q; + + if (!qdisc_dev(root)) + return (root->handle == handle ? root : NULL); + + if (!(root->flags & TCQ_F_BUILTIN) && + root->handle == handle) + return root; + + hash_for_each_possible_rcu(qdisc_dev(root)->qdisc_hash, q, hash, handle) { + if (q->handle == handle) + return q; + } + return NULL; +} + +void qdisc_hash_add(struct Qdisc *q, bool invisible) +{ + if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { + ASSERT_RTNL(); + hash_add_rcu(qdisc_dev(q)->qdisc_hash, &q->hash, q->handle); + if (invisible) + q->flags |= TCQ_F_INVISIBLE; + } +} +EXPORT_SYMBOL(qdisc_hash_add); + +void qdisc_hash_del(struct Qdisc *q) +{ + if ((q->parent != TC_H_ROOT) && !(q->flags & TCQ_F_INGRESS)) { + ASSERT_RTNL(); + hash_del_rcu(&q->hash); + } +} +EXPORT_SYMBOL(qdisc_hash_del); + +struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle) +{ + struct Qdisc *q; + + if (!handle) + return NULL; + q = qdisc_match_from_root(dev->qdisc, handle); + if (q) + goto out; + + if (dev_ingress_queue(dev)) + q = qdisc_match_from_root( + dev_ingress_queue(dev)->qdisc_sleeping, + handle); +out: + return q; +} + +struct Qdisc *qdisc_lookup_rcu(struct net_device *dev, u32 handle) +{ + struct netdev_queue *nq; + struct Qdisc *q; + + if (!handle) + return NULL; + q = qdisc_match_from_root(dev->qdisc, handle); + if (q) + goto out; + + nq = dev_ingress_queue_rcu(dev); + if (nq) + q = qdisc_match_from_root(nq->qdisc_sleeping, handle); +out: + return q; +} + +static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid) +{ + unsigned long cl; + struct Qdisc *leaf; + const struct Qdisc_class_ops *cops = p->ops->cl_ops; + + if (cops == NULL) + return NULL; + cl = cops->find(p, classid); + + if (cl == 0) + return NULL; + leaf = cops->leaf(p, cl); + return leaf; +} + +/* Find queueing discipline by name */ + +static struct Qdisc_ops *qdisc_lookup_ops(struct nlattr *kind) +{ + struct Qdisc_ops *q = NULL; + + if (kind) { + read_lock(&qdisc_mod_lock); + for (q = qdisc_base; q; q = q->next) { + if (nla_strcmp(kind, q->id) == 0) { + if (!try_module_get(q->owner)) + q = NULL; + break; + } + } + read_unlock(&qdisc_mod_lock); + } + return q; +} + +/* The linklayer setting were not transferred from iproute2, in older + * versions, and the rate tables lookup systems have been dropped in + * the kernel. To keep backward compatible with older iproute2 tc + * utils, we detect the linklayer setting by detecting if the rate + * table were modified. + * + * For linklayer ATM table entries, the rate table will be aligned to + * 48 bytes, thus some table entries will contain the same value. The + * mpu (min packet unit) is also encoded into the old rate table, thus + * starting from the mpu, we find low and high table entries for + * mapping this cell. If these entries contain the same value, when + * the rate tables have been modified for linklayer ATM. + * + * This is done by rounding mpu to the nearest 48 bytes cell/entry, + * and then roundup to the next cell, calc the table entry one below, + * and compare. + */ +static __u8 __detect_linklayer(struct tc_ratespec *r, __u32 *rtab) +{ + int low = roundup(r->mpu, 48); + int high = roundup(low+1, 48); + int cell_low = low >> r->cell_log; + int cell_high = (high >> r->cell_log) - 1; + + /* rtab is too inaccurate at rates > 100Mbit/s */ + if ((r->rate > (100000000/8)) || (rtab[0] == 0)) { + pr_debug("TC linklayer: Giving up ATM detection\n"); + return TC_LINKLAYER_ETHERNET; + } + + if ((cell_high > cell_low) && (cell_high < 256) + && (rtab[cell_low] == rtab[cell_high])) { + pr_debug("TC linklayer: Detected ATM, low(%d)=high(%d)=%u\n", + cell_low, cell_high, rtab[cell_high]); + return TC_LINKLAYER_ATM; + } + return TC_LINKLAYER_ETHERNET; +} + +static struct qdisc_rate_table *qdisc_rtab_list; + +struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, + struct nlattr *tab, + struct netlink_ext_ack *extack) +{ + struct qdisc_rate_table *rtab; + + if (tab == NULL || r->rate == 0 || + r->cell_log == 0 || r->cell_log >= 32 || + nla_len(tab) != TC_RTAB_SIZE) { + NL_SET_ERR_MSG(extack, "Invalid rate table parameters for searching"); + return NULL; + } + + for (rtab = qdisc_rtab_list; rtab; rtab = rtab->next) { + if (!memcmp(&rtab->rate, r, sizeof(struct tc_ratespec)) && + !memcmp(&rtab->data, nla_data(tab), 1024)) { + rtab->refcnt++; + return rtab; + } + } + + rtab = kmalloc(sizeof(*rtab), GFP_KERNEL); + if (rtab) { + rtab->rate = *r; + rtab->refcnt = 1; + memcpy(rtab->data, nla_data(tab), 1024); + if (r->linklayer == TC_LINKLAYER_UNAWARE) + r->linklayer = __detect_linklayer(r, rtab->data); + rtab->next = qdisc_rtab_list; + qdisc_rtab_list = rtab; + } else { + NL_SET_ERR_MSG(extack, "Failed to allocate new qdisc rate table"); + } + return rtab; +} +EXPORT_SYMBOL(qdisc_get_rtab); + +void qdisc_put_rtab(struct qdisc_rate_table *tab) +{ + struct qdisc_rate_table *rtab, **rtabp; + + if (!tab || --tab->refcnt) + return; + + for (rtabp = &qdisc_rtab_list; + (rtab = *rtabp) != NULL; + rtabp = &rtab->next) { + if (rtab == tab) { + *rtabp = rtab->next; + kfree(rtab); + return; + } + } +} +EXPORT_SYMBOL(qdisc_put_rtab); + +static LIST_HEAD(qdisc_stab_list); + +static const struct nla_policy stab_policy[TCA_STAB_MAX + 1] = { + [TCA_STAB_BASE] = { .len = sizeof(struct tc_sizespec) }, + [TCA_STAB_DATA] = { .type = NLA_BINARY }, +}; + +static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_STAB_MAX + 1]; + struct qdisc_size_table *stab; + struct tc_sizespec *s; + unsigned int tsize = 0; + u16 *tab = NULL; + int err; + + err = nla_parse_nested(tb, TCA_STAB_MAX, opt, stab_policy, extack); + if (err < 0) + return ERR_PTR(err); + if (!tb[TCA_STAB_BASE]) { + NL_SET_ERR_MSG(extack, "Size table base attribute is missing"); + return ERR_PTR(-EINVAL); + } + + s = nla_data(tb[TCA_STAB_BASE]); + + if (s->tsize > 0) { + if (!tb[TCA_STAB_DATA]) { + NL_SET_ERR_MSG(extack, "Size table data attribute is missing"); + return ERR_PTR(-EINVAL); + } + tab = nla_data(tb[TCA_STAB_DATA]); + tsize = nla_len(tb[TCA_STAB_DATA]) / sizeof(u16); + } + + if (tsize != s->tsize || (!tab && tsize > 0)) { + NL_SET_ERR_MSG(extack, "Invalid size of size table"); + return ERR_PTR(-EINVAL); + } + + list_for_each_entry(stab, &qdisc_stab_list, list) { + if (memcmp(&stab->szopts, s, sizeof(*s))) + continue; + if (tsize > 0 && memcmp(stab->data, tab, tsize * sizeof(u16))) + continue; + stab->refcnt++; + return stab; + } + + if (s->size_log > STAB_SIZE_LOG_MAX || + s->cell_log > STAB_SIZE_LOG_MAX) { + NL_SET_ERR_MSG(extack, "Invalid logarithmic size of size table"); + return ERR_PTR(-EINVAL); + } + + stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL); + if (!stab) + return ERR_PTR(-ENOMEM); + + stab->refcnt = 1; + stab->szopts = *s; + if (tsize > 0) + memcpy(stab->data, tab, tsize * sizeof(u16)); + + list_add_tail(&stab->list, &qdisc_stab_list); + + return stab; +} + +static void stab_kfree_rcu(struct rcu_head *head) +{ + kfree(container_of(head, struct qdisc_size_table, rcu)); +} + +void qdisc_put_stab(struct qdisc_size_table *tab) +{ + if (!tab) + return; + + if (--tab->refcnt == 0) { + list_del(&tab->list); + call_rcu_bh(&tab->rcu, stab_kfree_rcu); + } +} +EXPORT_SYMBOL(qdisc_put_stab); + +static int qdisc_dump_stab(struct sk_buff *skb, struct qdisc_size_table *stab) +{ + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_STAB); + if (nest == NULL) + goto nla_put_failure; + if (nla_put(skb, TCA_STAB_BASE, sizeof(stab->szopts), &stab->szopts)) + goto nla_put_failure; + nla_nest_end(skb, nest); + + return skb->len; + +nla_put_failure: + return -1; +} + +void __qdisc_calculate_pkt_len(struct sk_buff *skb, + const struct qdisc_size_table *stab) +{ + int pkt_len, slot; + + pkt_len = skb->len + stab->szopts.overhead; + if (unlikely(!stab->szopts.tsize)) + goto out; + + slot = pkt_len + stab->szopts.cell_align; + if (unlikely(slot < 0)) + slot = 0; + + slot >>= stab->szopts.cell_log; + if (likely(slot < stab->szopts.tsize)) + pkt_len = stab->data[slot]; + else + pkt_len = stab->data[stab->szopts.tsize - 1] * + (slot / stab->szopts.tsize) + + stab->data[slot % stab->szopts.tsize]; + + pkt_len <<= stab->szopts.size_log; +out: + if (unlikely(pkt_len < 1)) + pkt_len = 1; + qdisc_skb_cb(skb)->pkt_len = pkt_len; +} +EXPORT_SYMBOL(__qdisc_calculate_pkt_len); + +void qdisc_warn_nonwc(const char *txt, struct Qdisc *qdisc) +{ + if (!(qdisc->flags & TCQ_F_WARN_NONWC)) { + pr_warn("%s: %s qdisc %X: is non-work-conserving?\n", + txt, qdisc->ops->id, qdisc->handle >> 16); + qdisc->flags |= TCQ_F_WARN_NONWC; + } +} +EXPORT_SYMBOL(qdisc_warn_nonwc); + +static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer) +{ + struct qdisc_watchdog *wd = container_of(timer, struct qdisc_watchdog, + timer); + + rcu_read_lock(); + __netif_schedule(qdisc_root(wd->qdisc)); + rcu_read_unlock(); + + return HRTIMER_NORESTART; +} + +void qdisc_watchdog_init_clockid(struct qdisc_watchdog *wd, struct Qdisc *qdisc, + clockid_t clockid) +{ + hrtimer_init(&wd->timer, clockid, HRTIMER_MODE_ABS_PINNED); + wd->timer.function = qdisc_watchdog; + wd->qdisc = qdisc; +} +EXPORT_SYMBOL(qdisc_watchdog_init_clockid); + +void qdisc_watchdog_init(struct qdisc_watchdog *wd, struct Qdisc *qdisc) +{ + qdisc_watchdog_init_clockid(wd, qdisc, CLOCK_MONOTONIC); +} +EXPORT_SYMBOL(qdisc_watchdog_init); + +void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires) +{ + if (test_bit(__QDISC_STATE_DEACTIVATED, + &qdisc_root_sleeping(wd->qdisc)->state)) + return; + + if (wd->last_expires == expires) + return; + + wd->last_expires = expires; + hrtimer_start(&wd->timer, + ns_to_ktime(expires), + HRTIMER_MODE_ABS_PINNED); +} +EXPORT_SYMBOL(qdisc_watchdog_schedule_ns); + +void qdisc_watchdog_cancel(struct qdisc_watchdog *wd) +{ + hrtimer_cancel(&wd->timer); +} +EXPORT_SYMBOL(qdisc_watchdog_cancel); + +static struct hlist_head *qdisc_class_hash_alloc(unsigned int n) +{ + struct hlist_head *h; + unsigned int i; + + h = kvmalloc_array(n, sizeof(struct hlist_head), GFP_KERNEL); + + if (h != NULL) { + for (i = 0; i < n; i++) + INIT_HLIST_HEAD(&h[i]); + } + return h; +} + +void qdisc_class_hash_grow(struct Qdisc *sch, struct Qdisc_class_hash *clhash) +{ + struct Qdisc_class_common *cl; + struct hlist_node *next; + struct hlist_head *nhash, *ohash; + unsigned int nsize, nmask, osize; + unsigned int i, h; + + /* Rehash when load factor exceeds 0.75 */ + if (clhash->hashelems * 4 <= clhash->hashsize * 3) + return; + nsize = clhash->hashsize * 2; + nmask = nsize - 1; + nhash = qdisc_class_hash_alloc(nsize); + if (nhash == NULL) + return; + + ohash = clhash->hash; + osize = clhash->hashsize; + + sch_tree_lock(sch); + for (i = 0; i < osize; i++) { + hlist_for_each_entry_safe(cl, next, &ohash[i], hnode) { + h = qdisc_class_hash(cl->classid, nmask); + hlist_add_head(&cl->hnode, &nhash[h]); + } + } + clhash->hash = nhash; + clhash->hashsize = nsize; + clhash->hashmask = nmask; + sch_tree_unlock(sch); + + kvfree(ohash); +} +EXPORT_SYMBOL(qdisc_class_hash_grow); + +int qdisc_class_hash_init(struct Qdisc_class_hash *clhash) +{ + unsigned int size = 4; + + clhash->hash = qdisc_class_hash_alloc(size); + if (!clhash->hash) + return -ENOMEM; + clhash->hashsize = size; + clhash->hashmask = size - 1; + clhash->hashelems = 0; + return 0; +} +EXPORT_SYMBOL(qdisc_class_hash_init); + +void qdisc_class_hash_destroy(struct Qdisc_class_hash *clhash) +{ + kvfree(clhash->hash); +} +EXPORT_SYMBOL(qdisc_class_hash_destroy); + +void qdisc_class_hash_insert(struct Qdisc_class_hash *clhash, + struct Qdisc_class_common *cl) +{ + unsigned int h; + + INIT_HLIST_NODE(&cl->hnode); + h = qdisc_class_hash(cl->classid, clhash->hashmask); + hlist_add_head(&cl->hnode, &clhash->hash[h]); + clhash->hashelems++; +} +EXPORT_SYMBOL(qdisc_class_hash_insert); + +void qdisc_class_hash_remove(struct Qdisc_class_hash *clhash, + struct Qdisc_class_common *cl) +{ + hlist_del(&cl->hnode); + clhash->hashelems--; +} +EXPORT_SYMBOL(qdisc_class_hash_remove); + +/* Allocate an unique handle from space managed by kernel + * Possible range is [8000-FFFF]:0000 (0x8000 values) + */ +static u32 qdisc_alloc_handle(struct net_device *dev) +{ + int i = 0x8000; + static u32 autohandle = TC_H_MAKE(0x80000000U, 0); + + do { + autohandle += TC_H_MAKE(0x10000U, 0); + if (autohandle == TC_H_MAKE(TC_H_ROOT, 0)) + autohandle = TC_H_MAKE(0x80000000U, 0); + if (!qdisc_lookup(dev, autohandle)) + return autohandle; + cond_resched(); + } while (--i > 0); + + return 0; +} + +void qdisc_tree_reduce_backlog(struct Qdisc *sch, unsigned int n, + unsigned int len) +{ + bool qdisc_is_offloaded = sch->flags & TCQ_F_OFFLOADED; + const struct Qdisc_class_ops *cops; + unsigned long cl; + u32 parentid; + bool notify; + int drops; + + if (n == 0 && len == 0) + return; + drops = max_t(int, n, 0); + rcu_read_lock(); + while ((parentid = sch->parent)) { + if (TC_H_MAJ(parentid) == TC_H_MAJ(TC_H_INGRESS)) + break; + + if (sch->flags & TCQ_F_NOPARENT) + break; + /* Notify parent qdisc only if child qdisc becomes empty. + * + * If child was empty even before update then backlog + * counter is screwed and we skip notification because + * parent class is already passive. + * + * If the original child was offloaded then it is allowed + * to be seem as empty, so the parent is notified anyway. + */ + notify = !sch->q.qlen && !WARN_ON_ONCE(!n && + !qdisc_is_offloaded); + /* TODO: perform the search on a per txq basis */ + sch = qdisc_lookup(qdisc_dev(sch), TC_H_MAJ(parentid)); + if (sch == NULL) { + WARN_ON_ONCE(parentid != TC_H_ROOT); + break; + } + cops = sch->ops->cl_ops; + if (notify && cops->qlen_notify) { + cl = cops->find(sch, parentid); + cops->qlen_notify(sch, cl); + } + sch->q.qlen -= n; + sch->qstats.backlog -= len; + __qdisc_qstats_drop(sch, drops); + } + rcu_read_unlock(); +} +EXPORT_SYMBOL(qdisc_tree_reduce_backlog); + +static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, + u32 portid, u32 seq, u16 flags, int event) +{ + struct gnet_stats_basic_cpu __percpu *cpu_bstats = NULL; + struct gnet_stats_queue __percpu *cpu_qstats = NULL; + struct tcmsg *tcm; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + struct gnet_dump d; + struct qdisc_size_table *stab; + u32 block_index; + __u32 qlen; + + cond_resched(); + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + tcm->tcm_ifindex = qdisc_dev(q)->ifindex; + tcm->tcm_parent = clid; + tcm->tcm_handle = q->handle; + tcm->tcm_info = refcount_read(&q->refcnt); + if (nla_put_string(skb, TCA_KIND, q->ops->id)) + goto nla_put_failure; + if (q->ops->ingress_block_get) { + block_index = q->ops->ingress_block_get(q); + if (block_index && + nla_put_u32(skb, TCA_INGRESS_BLOCK, block_index)) + goto nla_put_failure; + } + if (q->ops->egress_block_get) { + block_index = q->ops->egress_block_get(q); + if (block_index && + nla_put_u32(skb, TCA_EGRESS_BLOCK, block_index)) + goto nla_put_failure; + } + if (q->ops->dump && q->ops->dump(q, skb) < 0) + goto nla_put_failure; + if (nla_put_u8(skb, TCA_HW_OFFLOAD, !!(q->flags & TCQ_F_OFFLOADED))) + goto nla_put_failure; + qlen = qdisc_qlen_sum(q); + + stab = rtnl_dereference(q->stab); + if (stab && qdisc_dump_stab(skb, stab) < 0) + goto nla_put_failure; + + if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, + NULL, &d, TCA_PAD) < 0) + goto nla_put_failure; + + if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0) + goto nla_put_failure; + + if (qdisc_is_percpu_stats(q)) { + cpu_bstats = q->cpu_bstats; + cpu_qstats = q->cpu_qstats; + } + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(q), + &d, cpu_bstats, &q->bstats) < 0 || + gnet_stats_copy_rate_est(&d, &q->rate_est) < 0 || + gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0) + goto nla_put_failure; + + if (gnet_stats_finish_copy(&d) < 0) + goto nla_put_failure; + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static bool tc_qdisc_dump_ignore(struct Qdisc *q, bool dump_invisible) +{ + if (q->flags & TCQ_F_BUILTIN) + return true; + if ((q->flags & TCQ_F_INVISIBLE) && !dump_invisible) + return true; + + return false; +} + +static int qdisc_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, u32 clid, + struct Qdisc *old, struct Qdisc *new) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (old && !tc_qdisc_dump_ignore(old, false)) { + if (tc_fill_qdisc(skb, old, clid, portid, n->nlmsg_seq, + 0, RTM_DELQDISC) < 0) + goto err_out; + } + if (new && !tc_qdisc_dump_ignore(new, false)) { + if (tc_fill_qdisc(skb, new, clid, portid, n->nlmsg_seq, + old ? NLM_F_REPLACE : 0, RTM_NEWQDISC) < 0) + goto err_out; + } + + if (skb->len) + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); + +err_out: + kfree_skb(skb); + return -EINVAL; +} + +static void notify_and_destroy(struct net *net, struct sk_buff *skb, + struct nlmsghdr *n, u32 clid, + struct Qdisc *old, struct Qdisc *new) +{ + if (new || old) + qdisc_notify(net, skb, n, clid, old, new); + + if (old) + qdisc_put(old); +} + +/* Graft qdisc "new" to class "classid" of qdisc "parent" or + * to device "dev". + * + * When appropriate send a netlink notification using 'skb' + * and "n". + * + * On success, destroy old qdisc. + */ + +static int qdisc_graft(struct net_device *dev, struct Qdisc *parent, + struct sk_buff *skb, struct nlmsghdr *n, u32 classid, + struct Qdisc *new, struct Qdisc *old, + struct netlink_ext_ack *extack) +{ + struct Qdisc *q = old; + struct net *net = dev_net(dev); + int err = 0; + + if (parent == NULL) { + unsigned int i, num_q, ingress; + + ingress = 0; + num_q = dev->num_tx_queues; + if ((q && q->flags & TCQ_F_INGRESS) || + (new && new->flags & TCQ_F_INGRESS)) { + num_q = 1; + ingress = 1; + if (!dev_ingress_queue(dev)) { + NL_SET_ERR_MSG(extack, "Device does not have an ingress queue"); + return -ENOENT; + } + } + + if (dev->flags & IFF_UP) + dev_deactivate(dev); + + if (new && new->ops->attach) + goto skip; + + for (i = 0; i < num_q; i++) { + struct netdev_queue *dev_queue = dev_ingress_queue(dev); + + if (!ingress) + dev_queue = netdev_get_tx_queue(dev, i); + + old = dev_graft_qdisc(dev_queue, new); + if (new && i > 0) + qdisc_refcount_inc(new); + + if (!ingress) + qdisc_put(old); + } + +skip: + if (!ingress) { + notify_and_destroy(net, skb, n, classid, + dev->qdisc, new); + if (new && !new->ops->attach) + qdisc_refcount_inc(new); + dev->qdisc = new ? : &noop_qdisc; + + if (new && new->ops->attach) + new->ops->attach(new); + } else { + notify_and_destroy(net, skb, n, classid, old, new); + } + + if (dev->flags & IFF_UP) + dev_activate(dev); + } else { + const struct Qdisc_class_ops *cops = parent->ops->cl_ops; + + /* Only support running class lockless if parent is lockless */ + if (new && (new->flags & TCQ_F_NOLOCK) && + parent && !(parent->flags & TCQ_F_NOLOCK)) + new->flags &= ~TCQ_F_NOLOCK; + + err = -EOPNOTSUPP; + if (cops && cops->graft) { + unsigned long cl = cops->find(parent, classid); + + if (cl) { + err = cops->graft(parent, cl, new, &old, + extack); + } else { + NL_SET_ERR_MSG(extack, "Specified class not found"); + err = -ENOENT; + } + } + if (!err) + notify_and_destroy(net, skb, n, classid, old, new); + } + return err; +} + +static int qdisc_block_indexes_set(struct Qdisc *sch, struct nlattr **tca, + struct netlink_ext_ack *extack) +{ + u32 block_index; + + if (tca[TCA_INGRESS_BLOCK]) { + block_index = nla_get_u32(tca[TCA_INGRESS_BLOCK]); + + if (!block_index) { + NL_SET_ERR_MSG(extack, "Ingress block index cannot be 0"); + return -EINVAL; + } + if (!sch->ops->ingress_block_set) { + NL_SET_ERR_MSG(extack, "Ingress block sharing is not supported"); + return -EOPNOTSUPP; + } + sch->ops->ingress_block_set(sch, block_index); + } + if (tca[TCA_EGRESS_BLOCK]) { + block_index = nla_get_u32(tca[TCA_EGRESS_BLOCK]); + + if (!block_index) { + NL_SET_ERR_MSG(extack, "Egress block index cannot be 0"); + return -EINVAL; + } + if (!sch->ops->egress_block_set) { + NL_SET_ERR_MSG(extack, "Egress block sharing is not supported"); + return -EOPNOTSUPP; + } + sch->ops->egress_block_set(sch, block_index); + } + return 0; +} + +/* lockdep annotation is needed for ingress; egress gets it only for name */ +static struct lock_class_key qdisc_tx_lock; +static struct lock_class_key qdisc_rx_lock; + +/* + Allocate and initialize new qdisc. + + Parameters are passed via opt. + */ + +static struct Qdisc *qdisc_create(struct net_device *dev, + struct netdev_queue *dev_queue, + struct Qdisc *p, u32 parent, u32 handle, + struct nlattr **tca, int *errp, + struct netlink_ext_ack *extack) +{ + int err; + struct nlattr *kind = tca[TCA_KIND]; + struct Qdisc *sch; + struct Qdisc_ops *ops; + struct qdisc_size_table *stab; + + ops = qdisc_lookup_ops(kind); +#ifdef CONFIG_MODULES + if (ops == NULL && kind != NULL) { + char name[IFNAMSIZ]; + if (nla_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) { + /* We dropped the RTNL semaphore in order to + * perform the module load. So, even if we + * succeeded in loading the module we have to + * tell the caller to replay the request. We + * indicate this using -EAGAIN. + * We replay the request because the device may + * go away in the mean time. + */ + rtnl_unlock(); + request_module("sch_%s", name); + rtnl_lock(); + ops = qdisc_lookup_ops(kind); + if (ops != NULL) { + /* We will try again qdisc_lookup_ops, + * so don't keep a reference. + */ + module_put(ops->owner); + err = -EAGAIN; + goto err_out; + } + } + } +#endif + + err = -ENOENT; + if (!ops) { + NL_SET_ERR_MSG(extack, "Specified qdisc kind is unknown"); + goto err_out; + } + + sch = qdisc_alloc(dev_queue, ops, extack); + if (IS_ERR(sch)) { + err = PTR_ERR(sch); + goto err_out2; + } + + sch->parent = parent; + + if (handle == TC_H_INGRESS) { + sch->flags |= TCQ_F_INGRESS; + handle = TC_H_MAKE(TC_H_INGRESS, 0); + lockdep_set_class(qdisc_lock(sch), &qdisc_rx_lock); + } else { + if (handle == 0) { + handle = qdisc_alloc_handle(dev); + err = -ENOMEM; + if (handle == 0) + goto err_out3; + } + lockdep_set_class(qdisc_lock(sch), &qdisc_tx_lock); + if (!netif_is_multiqueue(dev)) + sch->flags |= TCQ_F_ONETXQUEUE; + } + + sch->handle = handle; + + /* This exist to keep backward compatible with a userspace + * loophole, what allowed userspace to get IFF_NO_QUEUE + * facility on older kernels by setting tx_queue_len=0 (prior + * to qdisc init), and then forgot to reinit tx_queue_len + * before again attaching a qdisc. + */ + if ((dev->priv_flags & IFF_NO_QUEUE) && (dev->tx_queue_len == 0)) { + dev->tx_queue_len = DEFAULT_TX_QUEUE_LEN; + netdev_info(dev, "Caught tx_queue_len zero misconfig\n"); + } + + err = qdisc_block_indexes_set(sch, tca, extack); + if (err) + goto err_out3; + + if (ops->init) { + err = ops->init(sch, tca[TCA_OPTIONS], extack); + if (err != 0) + goto err_out5; + } + + if (tca[TCA_STAB]) { + stab = qdisc_get_stab(tca[TCA_STAB], extack); + if (IS_ERR(stab)) { + err = PTR_ERR(stab); + goto err_out4; + } + rcu_assign_pointer(sch->stab, stab); + } + if (tca[TCA_RATE]) { + seqcount_t *running; + + err = -EOPNOTSUPP; + if (sch->flags & TCQ_F_MQROOT) { + NL_SET_ERR_MSG(extack, "Cannot attach rate estimator to a multi-queue root qdisc"); + goto err_out4; + } + + if (sch->parent != TC_H_ROOT && + !(sch->flags & TCQ_F_INGRESS) && + (!p || !(p->flags & TCQ_F_MQROOT))) + running = qdisc_root_sleeping_running(sch); + else + running = &sch->running; + + err = gen_new_estimator(&sch->bstats, + sch->cpu_bstats, + &sch->rate_est, + NULL, + running, + tca[TCA_RATE]); + if (err) { + NL_SET_ERR_MSG(extack, "Failed to generate new estimator"); + goto err_out4; + } + } + + qdisc_hash_add(sch, false); + + return sch; + +err_out5: + /* ops->init() failed, we call ->destroy() like qdisc_create_dflt() */ + if (ops->destroy) + ops->destroy(sch); +err_out3: + dev_put(dev); + qdisc_free(sch); +err_out2: + module_put(ops->owner); +err_out: + *errp = err; + return NULL; + +err_out4: + /* + * Any broken qdiscs that would require a ops->reset() here? + * The qdisc was never in action so it shouldn't be necessary. + */ + qdisc_put_stab(rtnl_dereference(sch->stab)); + if (ops->destroy) + ops->destroy(sch); + goto err_out3; +} + +static int qdisc_change(struct Qdisc *sch, struct nlattr **tca, + struct netlink_ext_ack *extack) +{ + struct qdisc_size_table *ostab, *stab = NULL; + int err = 0; + + if (tca[TCA_OPTIONS]) { + if (!sch->ops->change) { + NL_SET_ERR_MSG(extack, "Change operation not supported by specified qdisc"); + return -EINVAL; + } + if (tca[TCA_INGRESS_BLOCK] || tca[TCA_EGRESS_BLOCK]) { + NL_SET_ERR_MSG(extack, "Change of blocks is not supported"); + return -EOPNOTSUPP; + } + err = sch->ops->change(sch, tca[TCA_OPTIONS], extack); + if (err) + return err; + } + + if (tca[TCA_STAB]) { + stab = qdisc_get_stab(tca[TCA_STAB], extack); + if (IS_ERR(stab)) + return PTR_ERR(stab); + } + + ostab = rtnl_dereference(sch->stab); + rcu_assign_pointer(sch->stab, stab); + qdisc_put_stab(ostab); + + if (tca[TCA_RATE]) { + /* NB: ignores errors from replace_estimator + because change can't be undone. */ + if (sch->flags & TCQ_F_MQROOT) + goto out; + gen_replace_estimator(&sch->bstats, + sch->cpu_bstats, + &sch->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + } +out: + return 0; +} + +struct check_loop_arg { + struct qdisc_walker w; + struct Qdisc *p; + int depth; +}; + +static int check_loop_fn(struct Qdisc *q, unsigned long cl, + struct qdisc_walker *w); + +static int check_loop(struct Qdisc *q, struct Qdisc *p, int depth) +{ + struct check_loop_arg arg; + + if (q->ops->cl_ops == NULL) + return 0; + + arg.w.stop = arg.w.skip = arg.w.count = 0; + arg.w.fn = check_loop_fn; + arg.depth = depth; + arg.p = p; + q->ops->cl_ops->walk(q, &arg.w); + return arg.w.stop ? -ELOOP : 0; +} + +static int +check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w) +{ + struct Qdisc *leaf; + const struct Qdisc_class_ops *cops = q->ops->cl_ops; + struct check_loop_arg *arg = (struct check_loop_arg *)w; + + leaf = cops->leaf(q, cl); + if (leaf) { + if (leaf == arg->p || arg->depth > 7) + return -ELOOP; + return check_loop(leaf, arg->p, arg->depth + 1); + } + return 0; +} + +const struct nla_policy rtm_tca_policy[TCA_MAX + 1] = { + [TCA_KIND] = { .type = NLA_NUL_STRING, + .len = IFNAMSIZ - 1 }, + [TCA_RATE] = { .type = NLA_BINARY, + .len = sizeof(struct tc_estimator) }, + [TCA_STAB] = { .type = NLA_NESTED }, + [TCA_DUMP_INVISIBLE] = { .type = NLA_FLAG }, + [TCA_CHAIN] = { .type = NLA_U32 }, + [TCA_INGRESS_BLOCK] = { .type = NLA_U32 }, + [TCA_EGRESS_BLOCK] = { .type = NLA_U32 }, +}; + +/* + * Delete/get qdisc. + */ + +static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct tcmsg *tcm = nlmsg_data(n); + struct nlattr *tca[TCA_MAX + 1]; + struct net_device *dev; + u32 clid; + struct Qdisc *q = NULL; + struct Qdisc *p = NULL; + int err; + + if ((n->nlmsg_type != RTM_GETQDISC) && + !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, + extack); + if (err < 0) + return err; + + dev = __dev_get_by_index(net, tcm->tcm_ifindex); + if (!dev) + return -ENODEV; + + clid = tcm->tcm_parent; + if (clid) { + if (clid != TC_H_ROOT) { + if (TC_H_MAJ(clid) != TC_H_MAJ(TC_H_INGRESS)) { + p = qdisc_lookup(dev, TC_H_MAJ(clid)); + if (!p) { + NL_SET_ERR_MSG(extack, "Failed to find qdisc with specified classid"); + return -ENOENT; + } + q = qdisc_leaf(p, clid); + } else if (dev_ingress_queue(dev)) { + q = dev_ingress_queue(dev)->qdisc_sleeping; + } + } else { + q = dev->qdisc; + } + if (!q) { + NL_SET_ERR_MSG(extack, "Cannot find specified qdisc on specified device"); + return -ENOENT; + } + + if (tcm->tcm_handle && q->handle != tcm->tcm_handle) { + NL_SET_ERR_MSG(extack, "Invalid handle"); + return -EINVAL; + } + } else { + q = qdisc_lookup(dev, tcm->tcm_handle); + if (!q) { + NL_SET_ERR_MSG(extack, "Failed to find qdisc with specified handle"); + return -ENOENT; + } + } + + if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id)) { + NL_SET_ERR_MSG(extack, "Invalid qdisc name"); + return -EINVAL; + } + + if (n->nlmsg_type == RTM_DELQDISC) { + if (!clid) { + NL_SET_ERR_MSG(extack, "Classid cannot be zero"); + return -EINVAL; + } + if (q->handle == 0) { + NL_SET_ERR_MSG(extack, "Cannot delete qdisc with handle of zero"); + return -ENOENT; + } + err = qdisc_graft(dev, p, skb, n, clid, NULL, q, extack); + if (err != 0) + return err; + } else { + qdisc_notify(net, skb, n, clid, NULL, q); + } + return 0; +} + +/* + * Create/change qdisc. + */ + +static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct tcmsg *tcm; + struct nlattr *tca[TCA_MAX + 1]; + struct net_device *dev; + u32 clid; + struct Qdisc *q, *p; + int err; + + if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + +replay: + /* Reinit, just in case something touches this. */ + err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, + extack); + if (err < 0) + return err; + + tcm = nlmsg_data(n); + clid = tcm->tcm_parent; + q = p = NULL; + + dev = __dev_get_by_index(net, tcm->tcm_ifindex); + if (!dev) + return -ENODEV; + + + if (clid) { + if (clid != TC_H_ROOT) { + if (clid != TC_H_INGRESS) { + p = qdisc_lookup(dev, TC_H_MAJ(clid)); + if (!p) { + NL_SET_ERR_MSG(extack, "Failed to find specified qdisc"); + return -ENOENT; + } + q = qdisc_leaf(p, clid); + } else if (dev_ingress_queue_create(dev)) { + q = dev_ingress_queue(dev)->qdisc_sleeping; + } + } else { + q = dev->qdisc; + } + + /* It may be default qdisc, ignore it */ + if (q && q->handle == 0) + q = NULL; + + if (!q || !tcm->tcm_handle || q->handle != tcm->tcm_handle) { + if (tcm->tcm_handle) { + if (q && !(n->nlmsg_flags & NLM_F_REPLACE)) { + NL_SET_ERR_MSG(extack, "NLM_F_REPLACE needed to override"); + return -EEXIST; + } + if (TC_H_MIN(tcm->tcm_handle)) { + NL_SET_ERR_MSG(extack, "Invalid minor handle"); + return -EINVAL; + } + q = qdisc_lookup(dev, tcm->tcm_handle); + if (!q) + goto create_n_graft; + if (n->nlmsg_flags & NLM_F_EXCL) { + NL_SET_ERR_MSG(extack, "Exclusivity flag on, cannot override"); + return -EEXIST; + } + if (tca[TCA_KIND] && + nla_strcmp(tca[TCA_KIND], q->ops->id)) { + NL_SET_ERR_MSG(extack, "Invalid qdisc name"); + return -EINVAL; + } + if (q == p || + (p && check_loop(q, p, 0))) { + NL_SET_ERR_MSG(extack, "Qdisc parent/child loop detected"); + return -ELOOP; + } + qdisc_refcount_inc(q); + goto graft; + } else { + if (!q) + goto create_n_graft; + + /* This magic test requires explanation. + * + * We know, that some child q is already + * attached to this parent and have choice: + * either to change it or to create/graft new one. + * + * 1. We are allowed to create/graft only + * if CREATE and REPLACE flags are set. + * + * 2. If EXCL is set, requestor wanted to say, + * that qdisc tcm_handle is not expected + * to exist, so that we choose create/graft too. + * + * 3. The last case is when no flags are set. + * Alas, it is sort of hole in API, we + * cannot decide what to do unambiguously. + * For now we select create/graft, if + * user gave KIND, which does not match existing. + */ + if ((n->nlmsg_flags & NLM_F_CREATE) && + (n->nlmsg_flags & NLM_F_REPLACE) && + ((n->nlmsg_flags & NLM_F_EXCL) || + (tca[TCA_KIND] && + nla_strcmp(tca[TCA_KIND], q->ops->id)))) + goto create_n_graft; + } + } + } else { + if (!tcm->tcm_handle) { + NL_SET_ERR_MSG(extack, "Handle cannot be zero"); + return -EINVAL; + } + q = qdisc_lookup(dev, tcm->tcm_handle); + } + + /* Change qdisc parameters */ + if (!q) { + NL_SET_ERR_MSG(extack, "Specified qdisc not found"); + return -ENOENT; + } + if (n->nlmsg_flags & NLM_F_EXCL) { + NL_SET_ERR_MSG(extack, "Exclusivity flag on, cannot modify"); + return -EEXIST; + } + if (tca[TCA_KIND] && nla_strcmp(tca[TCA_KIND], q->ops->id)) { + NL_SET_ERR_MSG(extack, "Invalid qdisc name"); + return -EINVAL; + } + err = qdisc_change(q, tca, extack); + if (err == 0) + qdisc_notify(net, skb, n, clid, NULL, q); + return err; + +create_n_graft: + if (!(n->nlmsg_flags & NLM_F_CREATE)) { + NL_SET_ERR_MSG(extack, "Qdisc not found. To create specify NLM_F_CREATE flag"); + return -ENOENT; + } + if (clid == TC_H_INGRESS) { + if (dev_ingress_queue(dev)) { + q = qdisc_create(dev, dev_ingress_queue(dev), p, + tcm->tcm_parent, tcm->tcm_parent, + tca, &err, extack); + } else { + NL_SET_ERR_MSG(extack, "Cannot find ingress queue for specified device"); + err = -ENOENT; + } + } else { + struct netdev_queue *dev_queue; + + if (p && p->ops->cl_ops && p->ops->cl_ops->select_queue) + dev_queue = p->ops->cl_ops->select_queue(p, tcm); + else if (p) + dev_queue = p->dev_queue; + else + dev_queue = netdev_get_tx_queue(dev, 0); + + q = qdisc_create(dev, dev_queue, p, + tcm->tcm_parent, tcm->tcm_handle, + tca, &err, extack); + } + if (q == NULL) { + if (err == -EAGAIN) + goto replay; + return err; + } + +graft: + err = qdisc_graft(dev, p, skb, n, clid, q, NULL, extack); + if (err) { + if (q) + qdisc_put(q); + return err; + } + + return 0; +} + +static int tc_dump_qdisc_root(struct Qdisc *root, struct sk_buff *skb, + struct netlink_callback *cb, + int *q_idx_p, int s_q_idx, bool recur, + bool dump_invisible) +{ + int ret = 0, q_idx = *q_idx_p; + struct Qdisc *q; + int b; + + if (!root) + return 0; + + q = root; + if (q_idx < s_q_idx) { + q_idx++; + } else { + if (!tc_qdisc_dump_ignore(q, dump_invisible) && + tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + RTM_NEWQDISC) <= 0) + goto done; + q_idx++; + } + + /* If dumping singletons, there is no qdisc_dev(root) and the singleton + * itself has already been dumped. + * + * If we've already dumped the top-level (ingress) qdisc above and the global + * qdisc hashtable, we don't want to hit it again + */ + if (!qdisc_dev(root) || !recur) + goto out; + + hash_for_each(qdisc_dev(root)->qdisc_hash, b, q, hash) { + if (q_idx < s_q_idx) { + q_idx++; + continue; + } + if (!tc_qdisc_dump_ignore(q, dump_invisible) && + tc_fill_qdisc(skb, q, q->parent, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + RTM_NEWQDISC) <= 0) + goto done; + q_idx++; + } + +out: + *q_idx_p = q_idx; + return ret; +done: + ret = -1; + goto out; +} + +static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + int idx, q_idx; + int s_idx, s_q_idx; + struct net_device *dev; + const struct nlmsghdr *nlh = cb->nlh; + struct nlattr *tca[TCA_MAX + 1]; + int err; + + s_idx = cb->args[0]; + s_q_idx = q_idx = cb->args[1]; + + idx = 0; + ASSERT_RTNL(); + + err = nlmsg_parse(nlh, sizeof(struct tcmsg), tca, TCA_MAX, + rtm_tca_policy, NULL); + if (err < 0) + return err; + + for_each_netdev(net, dev) { + struct netdev_queue *dev_queue; + + if (idx < s_idx) + goto cont; + if (idx > s_idx) + s_q_idx = 0; + q_idx = 0; + + if (tc_dump_qdisc_root(dev->qdisc, skb, cb, &q_idx, s_q_idx, + true, tca[TCA_DUMP_INVISIBLE]) < 0) + goto done; + + dev_queue = dev_ingress_queue(dev); + if (dev_queue && + tc_dump_qdisc_root(dev_queue->qdisc_sleeping, skb, cb, + &q_idx, s_q_idx, false, + tca[TCA_DUMP_INVISIBLE]) < 0) + goto done; + +cont: + idx++; + } + +done: + cb->args[0] = idx; + cb->args[1] = q_idx; + + return skb->len; +} + + + +/************************************************ + * Traffic classes manipulation. * + ************************************************/ + +static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, + unsigned long cl, + u32 portid, u32 seq, u16 flags, int event) +{ + struct tcmsg *tcm; + struct nlmsghdr *nlh; + unsigned char *b = skb_tail_pointer(skb); + struct gnet_dump d; + const struct Qdisc_class_ops *cl_ops = q->ops->cl_ops; + + cond_resched(); + nlh = nlmsg_put(skb, portid, seq, event, sizeof(*tcm), flags); + if (!nlh) + goto out_nlmsg_trim; + tcm = nlmsg_data(nlh); + tcm->tcm_family = AF_UNSPEC; + tcm->tcm__pad1 = 0; + tcm->tcm__pad2 = 0; + tcm->tcm_ifindex = qdisc_dev(q)->ifindex; + tcm->tcm_parent = q->handle; + tcm->tcm_handle = q->handle; + tcm->tcm_info = 0; + if (nla_put_string(skb, TCA_KIND, q->ops->id)) + goto nla_put_failure; + if (cl_ops->dump && cl_ops->dump(q, cl, skb, tcm) < 0) + goto nla_put_failure; + + if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, + NULL, &d, TCA_PAD) < 0) + goto nla_put_failure; + + if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0) + goto nla_put_failure; + + if (gnet_stats_finish_copy(&d) < 0) + goto nla_put_failure; + + nlh->nlmsg_len = skb_tail_pointer(skb) - b; + return skb->len; + +out_nlmsg_trim: +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int tclass_notify(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, struct Qdisc *q, + unsigned long cl, int event) +{ + struct sk_buff *skb; + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, event) < 0) { + kfree_skb(skb); + return -EINVAL; + } + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +static int tclass_del_notify(struct net *net, + const struct Qdisc_class_ops *cops, + struct sk_buff *oskb, struct nlmsghdr *n, + struct Qdisc *q, unsigned long cl) +{ + u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; + struct sk_buff *skb; + int err = 0; + + if (!cops->delete) + return -EOPNOTSUPP; + + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOBUFS; + + if (tc_fill_tclass(skb, q, cl, portid, n->nlmsg_seq, 0, + RTM_DELTCLASS) < 0) { + kfree_skb(skb); + return -EINVAL; + } + + err = cops->delete(q, cl); + if (err) { + kfree_skb(skb); + return err; + } + + return rtnetlink_send(skb, net, portid, RTNLGRP_TC, + n->nlmsg_flags & NLM_F_ECHO); +} + +#ifdef CONFIG_NET_CLS + +struct tcf_bind_args { + struct tcf_walker w; + unsigned long base; + unsigned long cl; + u32 classid; +}; + +static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg) +{ + struct tcf_bind_args *a = (void *)arg; + + if (tp->ops->bind_class) { + struct Qdisc *q = tcf_block_q(tp->chain->block); + + sch_tree_lock(q); + tp->ops->bind_class(n, a->classid, a->cl, q, a->base); + sch_tree_unlock(q); + } + return 0; +} + +static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid, + unsigned long new_cl) +{ + const struct Qdisc_class_ops *cops = q->ops->cl_ops; + struct tcf_block *block; + struct tcf_chain *chain; + unsigned long cl; + + cl = cops->find(q, portid); + if (!cl) + return; + if (!cops->tcf_block) + return; + block = cops->tcf_block(q, cl, NULL); + if (!block) + return; + list_for_each_entry(chain, &block->chain_list, list) { + struct tcf_proto *tp; + + for (tp = rtnl_dereference(chain->filter_chain); + tp; tp = rtnl_dereference(tp->next)) { + struct tcf_bind_args arg = {}; + + arg.w.fn = tcf_node_bind; + arg.classid = clid; + arg.base = cl; + arg.cl = new_cl; + tp->ops->walk(tp, &arg.w); + } + } +} + +#else + +static void tc_bind_tclass(struct Qdisc *q, u32 portid, u32 clid, + unsigned long new_cl) +{ +} + +#endif + +static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, + struct netlink_ext_ack *extack) +{ + struct net *net = sock_net(skb->sk); + struct tcmsg *tcm = nlmsg_data(n); + struct nlattr *tca[TCA_MAX + 1]; + struct net_device *dev; + struct Qdisc *q = NULL; + const struct Qdisc_class_ops *cops; + unsigned long cl = 0; + unsigned long new_cl; + u32 portid; + u32 clid; + u32 qid; + int err; + + if ((n->nlmsg_type != RTM_GETTCLASS) && + !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) + return -EPERM; + + err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, rtm_tca_policy, + extack); + if (err < 0) + return err; + + dev = __dev_get_by_index(net, tcm->tcm_ifindex); + if (!dev) + return -ENODEV; + + /* + parent == TC_H_UNSPEC - unspecified parent. + parent == TC_H_ROOT - class is root, which has no parent. + parent == X:0 - parent is root class. + parent == X:Y - parent is a node in hierarchy. + parent == 0:Y - parent is X:Y, where X:0 is qdisc. + + handle == 0:0 - generate handle from kernel pool. + handle == 0:Y - class is X:Y, where X:0 is qdisc. + handle == X:Y - clear. + handle == X:0 - root class. + */ + + /* Step 1. Determine qdisc handle X:0 */ + + portid = tcm->tcm_parent; + clid = tcm->tcm_handle; + qid = TC_H_MAJ(clid); + + if (portid != TC_H_ROOT) { + u32 qid1 = TC_H_MAJ(portid); + + if (qid && qid1) { + /* If both majors are known, they must be identical. */ + if (qid != qid1) + return -EINVAL; + } else if (qid1) { + qid = qid1; + } else if (qid == 0) + qid = dev->qdisc->handle; + + /* Now qid is genuine qdisc handle consistent + * both with parent and child. + * + * TC_H_MAJ(portid) still may be unspecified, complete it now. + */ + if (portid) + portid = TC_H_MAKE(qid, portid); + } else { + if (qid == 0) + qid = dev->qdisc->handle; + } + + /* OK. Locate qdisc */ + q = qdisc_lookup(dev, qid); + if (!q) + return -ENOENT; + + /* An check that it supports classes */ + cops = q->ops->cl_ops; + if (cops == NULL) + return -EINVAL; + + /* Now try to get class */ + if (clid == 0) { + if (portid == TC_H_ROOT) + clid = qid; + } else + clid = TC_H_MAKE(qid, clid); + + if (clid) + cl = cops->find(q, clid); + + if (cl == 0) { + err = -ENOENT; + if (n->nlmsg_type != RTM_NEWTCLASS || + !(n->nlmsg_flags & NLM_F_CREATE)) + goto out; + } else { + switch (n->nlmsg_type) { + case RTM_NEWTCLASS: + err = -EEXIST; + if (n->nlmsg_flags & NLM_F_EXCL) + goto out; + break; + case RTM_DELTCLASS: + err = tclass_del_notify(net, cops, skb, n, q, cl); + /* Unbind the class with flilters with 0 */ + tc_bind_tclass(q, portid, clid, 0); + goto out; + case RTM_GETTCLASS: + err = tclass_notify(net, skb, n, q, cl, RTM_NEWTCLASS); + goto out; + default: + err = -EINVAL; + goto out; + } + } + + if (tca[TCA_INGRESS_BLOCK] || tca[TCA_EGRESS_BLOCK]) { + NL_SET_ERR_MSG(extack, "Shared blocks are not supported for classes"); + return -EOPNOTSUPP; + } + + new_cl = cl; + err = -EOPNOTSUPP; + if (cops->change) + err = cops->change(q, clid, portid, tca, &new_cl, extack); + if (err == 0) { + tclass_notify(net, skb, n, q, new_cl, RTM_NEWTCLASS); + /* We just create a new class, need to do reverse binding. */ + if (cl != new_cl) + tc_bind_tclass(q, portid, clid, new_cl); + } +out: + return err; +} + +struct qdisc_dump_args { + struct qdisc_walker w; + struct sk_buff *skb; + struct netlink_callback *cb; +}; + +static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, + struct qdisc_walker *arg) +{ + struct qdisc_dump_args *a = (struct qdisc_dump_args *)arg; + + return tc_fill_tclass(a->skb, q, cl, NETLINK_CB(a->cb->skb).portid, + a->cb->nlh->nlmsg_seq, NLM_F_MULTI, + RTM_NEWTCLASS); +} + +static int tc_dump_tclass_qdisc(struct Qdisc *q, struct sk_buff *skb, + struct tcmsg *tcm, struct netlink_callback *cb, + int *t_p, int s_t) +{ + struct qdisc_dump_args arg; + + if (tc_qdisc_dump_ignore(q, false) || + *t_p < s_t || !q->ops->cl_ops || + (tcm->tcm_parent && + TC_H_MAJ(tcm->tcm_parent) != q->handle)) { + (*t_p)++; + return 0; + } + if (*t_p > s_t) + memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0])); + arg.w.fn = qdisc_class_dump; + arg.skb = skb; + arg.cb = cb; + arg.w.stop = 0; + arg.w.skip = cb->args[1]; + arg.w.count = 0; + q->ops->cl_ops->walk(q, &arg.w); + cb->args[1] = arg.w.count; + if (arg.w.stop) + return -1; + (*t_p)++; + return 0; +} + +static int tc_dump_tclass_root(struct Qdisc *root, struct sk_buff *skb, + struct tcmsg *tcm, struct netlink_callback *cb, + int *t_p, int s_t, bool recur) +{ + struct Qdisc *q; + int b; + + if (!root) + return 0; + + if (tc_dump_tclass_qdisc(root, skb, tcm, cb, t_p, s_t) < 0) + return -1; + + if (!qdisc_dev(root) || !recur) + return 0; + + if (tcm->tcm_parent) { + q = qdisc_match_from_root(root, TC_H_MAJ(tcm->tcm_parent)); + if (q && q != root && + tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0) + return -1; + return 0; + } + hash_for_each(qdisc_dev(root)->qdisc_hash, b, q, hash) { + if (tc_dump_tclass_qdisc(q, skb, tcm, cb, t_p, s_t) < 0) + return -1; + } + + return 0; +} + +static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct tcmsg *tcm = nlmsg_data(cb->nlh); + struct net *net = sock_net(skb->sk); + struct netdev_queue *dev_queue; + struct net_device *dev; + int t, s_t; + + if (nlmsg_len(cb->nlh) < sizeof(*tcm)) + return 0; + dev = dev_get_by_index(net, tcm->tcm_ifindex); + if (!dev) + return 0; + + s_t = cb->args[0]; + t = 0; + + if (tc_dump_tclass_root(dev->qdisc, skb, tcm, cb, &t, s_t, true) < 0) + goto done; + + dev_queue = dev_ingress_queue(dev); + if (dev_queue && + tc_dump_tclass_root(dev_queue->qdisc_sleeping, skb, tcm, cb, + &t, s_t, false) < 0) + goto done; + +done: + cb->args[0] = t; + + dev_put(dev); + return skb->len; +} + +#ifdef CONFIG_PROC_FS +static int psched_show(struct seq_file *seq, void *v) +{ + seq_printf(seq, "%08x %08x %08x %08x\n", + (u32)NSEC_PER_USEC, (u32)PSCHED_TICKS2NS(1), + 1000000, + (u32)NSEC_PER_SEC / hrtimer_resolution); + + return 0; +} + +static int __net_init psched_net_init(struct net *net) +{ + struct proc_dir_entry *e; + + e = proc_create_single("psched", 0, net->proc_net, psched_show); + if (e == NULL) + return -ENOMEM; + + return 0; +} + +static void __net_exit psched_net_exit(struct net *net) +{ + remove_proc_entry("psched", net->proc_net); +} +#else +static int __net_init psched_net_init(struct net *net) +{ + return 0; +} + +static void __net_exit psched_net_exit(struct net *net) +{ +} +#endif + +static struct pernet_operations psched_net_ops = { + .init = psched_net_init, + .exit = psched_net_exit, +}; + +static int __init pktsched_init(void) +{ + int err; + + err = register_pernet_subsys(&psched_net_ops); + if (err) { + pr_err("pktsched_init: " + "cannot initialize per netns operations\n"); + return err; + } + + register_qdisc(&pfifo_fast_ops); + register_qdisc(&pfifo_qdisc_ops); + register_qdisc(&bfifo_qdisc_ops); + register_qdisc(&pfifo_head_drop_qdisc_ops); + register_qdisc(&mq_qdisc_ops); + register_qdisc(&noqueue_qdisc_ops); + + rtnl_register(PF_UNSPEC, RTM_NEWQDISC, tc_modify_qdisc, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELQDISC, tc_get_qdisc, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETQDISC, tc_get_qdisc, tc_dump_qdisc, + 0); + rtnl_register(PF_UNSPEC, RTM_NEWTCLASS, tc_ctl_tclass, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_DELTCLASS, tc_ctl_tclass, NULL, 0); + rtnl_register(PF_UNSPEC, RTM_GETTCLASS, tc_ctl_tclass, tc_dump_tclass, + 0); + + return 0; +} + +subsys_initcall(pktsched_init); diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c new file mode 100644 index 000000000..9a1bfa13a --- /dev/null +++ b/net/sched/sch_atm.c @@ -0,0 +1,705 @@ +/* net/sched/sch_atm.c - ATM VC selection "queueing discipline" */ + +/* Written 1998-2000 by Werner Almesberger, EPFL ICA */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/atmdev.h> +#include <linux/atmclip.h> +#include <linux/rtnetlink.h> +#include <linux/file.h> /* for fput */ +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +/* + * The ATM queuing discipline provides a framework for invoking classifiers + * (aka "filters"), which in turn select classes of this queuing discipline. + * Each class maps the flow(s) it is handling to a given VC. Multiple classes + * may share the same VC. + * + * When creating a class, VCs are specified by passing the number of the open + * socket descriptor by which the calling process references the VC. The kernel + * keeps the VC open at least until all classes using it are removed. + * + * In this file, most functions are named atm_tc_* to avoid confusion with all + * the atm_* in net/atm. This naming convention differs from what's used in the + * rest of net/sched. + * + * Known bugs: + * - sometimes messes up the IP stack + * - any manipulations besides the few operations described in the README, are + * untested and likely to crash the system + * - should lock the flow while there is data in the queue (?) + */ + +#define VCC2FLOW(vcc) ((struct atm_flow_data *) ((vcc)->user_back)) + +struct atm_flow_data { + struct Qdisc_class_common common; + struct Qdisc *q; /* FIFO, TBF, etc. */ + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + struct atm_vcc *vcc; /* VCC; NULL if VCC is closed */ + void (*old_pop)(struct atm_vcc *vcc, + struct sk_buff *skb); /* chaining */ + struct atm_qdisc_data *parent; /* parent qdisc */ + struct socket *sock; /* for closing */ + int ref; /* reference count */ + struct gnet_stats_basic_packed bstats; + struct gnet_stats_queue qstats; + struct list_head list; + struct atm_flow_data *excess; /* flow for excess traffic; + NULL to set CLP instead */ + int hdr_len; + unsigned char hdr[0]; /* header data; MUST BE LAST */ +}; + +struct atm_qdisc_data { + struct atm_flow_data link; /* unclassified skbs go here */ + struct list_head flows; /* NB: "link" is also on this + list */ + struct tasklet_struct task; /* dequeue tasklet */ +}; + +/* ------------------------- Class/flow operations ------------------------- */ + +static inline struct atm_flow_data *lookup_flow(struct Qdisc *sch, u32 classid) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow; + + list_for_each_entry(flow, &p->flows, list) { + if (flow->common.classid == classid) + return flow; + } + return NULL; +} + +static int atm_tc_graft(struct Qdisc *sch, unsigned long arg, + struct Qdisc *new, struct Qdisc **old, + struct netlink_ext_ack *extack) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow = (struct atm_flow_data *)arg; + + pr_debug("atm_tc_graft(sch %p,[qdisc %p],flow %p,new %p,old %p)\n", + sch, p, flow, new, old); + if (list_empty(&flow->list)) + return -EINVAL; + if (!new) + new = &noop_qdisc; + *old = flow->q; + flow->q = new; + if (*old) + qdisc_reset(*old); + return 0; +} + +static struct Qdisc *atm_tc_leaf(struct Qdisc *sch, unsigned long cl) +{ + struct atm_flow_data *flow = (struct atm_flow_data *)cl; + + pr_debug("atm_tc_leaf(sch %p,flow %p)\n", sch, flow); + return flow ? flow->q : NULL; +} + +static unsigned long atm_tc_find(struct Qdisc *sch, u32 classid) +{ + struct atm_qdisc_data *p __maybe_unused = qdisc_priv(sch); + struct atm_flow_data *flow; + + pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", __func__, sch, p, classid); + flow = lookup_flow(sch, classid); + pr_debug("%s: flow %p\n", __func__, flow); + return (unsigned long)flow; +} + +static unsigned long atm_tc_bind_filter(struct Qdisc *sch, + unsigned long parent, u32 classid) +{ + struct atm_qdisc_data *p __maybe_unused = qdisc_priv(sch); + struct atm_flow_data *flow; + + pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", __func__, sch, p, classid); + flow = lookup_flow(sch, classid); + if (flow) + flow->ref++; + pr_debug("%s: flow %p\n", __func__, flow); + return (unsigned long)flow; +} + +/* + * atm_tc_put handles all destructions, including the ones that are explicitly + * requested (atm_tc_destroy, etc.). The assumption here is that we never drop + * anything that still seems to be in use. + */ +static void atm_tc_put(struct Qdisc *sch, unsigned long cl) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow = (struct atm_flow_data *)cl; + + pr_debug("atm_tc_put(sch %p,[qdisc %p],flow %p)\n", sch, p, flow); + if (--flow->ref) + return; + pr_debug("atm_tc_put: destroying\n"); + list_del_init(&flow->list); + pr_debug("atm_tc_put: qdisc %p\n", flow->q); + qdisc_put(flow->q); + tcf_block_put(flow->block); + if (flow->sock) { + pr_debug("atm_tc_put: f_count %ld\n", + file_count(flow->sock->file)); + flow->vcc->pop = flow->old_pop; + sockfd_put(flow->sock); + } + if (flow->excess) + atm_tc_put(sch, (unsigned long)flow->excess); + if (flow != &p->link) + kfree(flow); + /* + * If flow == &p->link, the qdisc no longer works at this point and + * needs to be removed. (By the caller of atm_tc_put.) + */ +} + +static void sch_atm_pop(struct atm_vcc *vcc, struct sk_buff *skb) +{ + struct atm_qdisc_data *p = VCC2FLOW(vcc)->parent; + + pr_debug("sch_atm_pop(vcc %p,skb %p,[qdisc %p])\n", vcc, skb, p); + VCC2FLOW(vcc)->old_pop(vcc, skb); + tasklet_schedule(&p->task); +} + +static const u8 llc_oui_ip[] = { + 0xaa, /* DSAP: non-ISO */ + 0xaa, /* SSAP: non-ISO */ + 0x03, /* Ctrl: Unnumbered Information Command PDU */ + 0x00, /* OUI: EtherType */ + 0x00, 0x00, + 0x08, 0x00 +}; /* Ethertype IP (0800) */ + +static const struct nla_policy atm_policy[TCA_ATM_MAX + 1] = { + [TCA_ATM_FD] = { .type = NLA_U32 }, + [TCA_ATM_EXCESS] = { .type = NLA_U32 }, +}; + +static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, + struct nlattr **tca, unsigned long *arg, + struct netlink_ext_ack *extack) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow = (struct atm_flow_data *)*arg; + struct atm_flow_data *excess = NULL; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_ATM_MAX + 1]; + struct socket *sock; + int fd, error, hdr_len; + void *hdr; + + pr_debug("atm_tc_change(sch %p,[qdisc %p],classid %x,parent %x," + "flow %p,opt %p)\n", sch, p, classid, parent, flow, opt); + /* + * The concept of parents doesn't apply for this qdisc. + */ + if (parent && parent != TC_H_ROOT && parent != sch->handle) + return -EINVAL; + /* + * ATM classes cannot be changed. In order to change properties of the + * ATM connection, that socket needs to be modified directly (via the + * native ATM API. In order to send a flow to a different VC, the old + * class needs to be removed and a new one added. (This may be changed + * later.) + */ + if (flow) + return -EBUSY; + if (opt == NULL) + return -EINVAL; + + error = nla_parse_nested(tb, TCA_ATM_MAX, opt, atm_policy, NULL); + if (error < 0) + return error; + + if (!tb[TCA_ATM_FD]) + return -EINVAL; + fd = nla_get_u32(tb[TCA_ATM_FD]); + pr_debug("atm_tc_change: fd %d\n", fd); + if (tb[TCA_ATM_HDR]) { + hdr_len = nla_len(tb[TCA_ATM_HDR]); + hdr = nla_data(tb[TCA_ATM_HDR]); + } else { + hdr_len = RFC1483LLC_LEN; + hdr = NULL; /* default LLC/SNAP for IP */ + } + if (!tb[TCA_ATM_EXCESS]) + excess = NULL; + else { + excess = (struct atm_flow_data *) + atm_tc_find(sch, nla_get_u32(tb[TCA_ATM_EXCESS])); + if (!excess) + return -ENOENT; + } + pr_debug("atm_tc_change: type %d, payload %d, hdr_len %d\n", + opt->nla_type, nla_len(opt), hdr_len); + sock = sockfd_lookup(fd, &error); + if (!sock) + return error; /* f_count++ */ + pr_debug("atm_tc_change: f_count %ld\n", file_count(sock->file)); + if (sock->ops->family != PF_ATMSVC && sock->ops->family != PF_ATMPVC) { + error = -EPROTOTYPE; + goto err_out; + } + /* @@@ should check if the socket is really operational or we'll crash + on vcc->send */ + if (classid) { + if (TC_H_MAJ(classid ^ sch->handle)) { + pr_debug("atm_tc_change: classid mismatch\n"); + error = -EINVAL; + goto err_out; + } + } else { + int i; + unsigned long cl; + + for (i = 1; i < 0x8000; i++) { + classid = TC_H_MAKE(sch->handle, 0x8000 | i); + cl = atm_tc_find(sch, classid); + if (!cl) + break; + } + } + pr_debug("atm_tc_change: new id %x\n", classid); + flow = kzalloc(sizeof(struct atm_flow_data) + hdr_len, GFP_KERNEL); + pr_debug("atm_tc_change: flow %p\n", flow); + if (!flow) { + error = -ENOBUFS; + goto err_out; + } + + error = tcf_block_get(&flow->block, &flow->filter_list, sch, + extack); + if (error) { + kfree(flow); + goto err_out; + } + + flow->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid, + extack); + if (!flow->q) + flow->q = &noop_qdisc; + pr_debug("atm_tc_change: qdisc %p\n", flow->q); + flow->sock = sock; + flow->vcc = ATM_SD(sock); /* speedup */ + flow->vcc->user_back = flow; + pr_debug("atm_tc_change: vcc %p\n", flow->vcc); + flow->old_pop = flow->vcc->pop; + flow->parent = p; + flow->vcc->pop = sch_atm_pop; + flow->common.classid = classid; + flow->ref = 1; + flow->excess = excess; + list_add(&flow->list, &p->link.list); + flow->hdr_len = hdr_len; + if (hdr) + memcpy(flow->hdr, hdr, hdr_len); + else + memcpy(flow->hdr, llc_oui_ip, sizeof(llc_oui_ip)); + *arg = (unsigned long)flow; + return 0; +err_out: + sockfd_put(sock); + return error; +} + +static int atm_tc_delete(struct Qdisc *sch, unsigned long arg) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow = (struct atm_flow_data *)arg; + + pr_debug("atm_tc_delete(sch %p,[qdisc %p],flow %p)\n", sch, p, flow); + if (list_empty(&flow->list)) + return -EINVAL; + if (rcu_access_pointer(flow->filter_list) || flow == &p->link) + return -EBUSY; + /* + * Reference count must be 2: one for "keepalive" (set at class + * creation), and one for the reference held when calling delete. + */ + if (flow->ref < 2) { + pr_err("atm_tc_delete: flow->ref == %d\n", flow->ref); + return -EINVAL; + } + if (flow->ref > 2) + return -EBUSY; /* catch references via excess, etc. */ + atm_tc_put(sch, arg); + return 0; +} + +static void atm_tc_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow; + + pr_debug("atm_tc_walk(sch %p,[qdisc %p],walker %p)\n", sch, p, walker); + if (walker->stop) + return; + list_for_each_entry(flow, &p->flows, list) { + if (walker->count >= walker->skip && + walker->fn(sch, (unsigned long)flow, walker) < 0) { + walker->stop = 1; + break; + } + walker->count++; + } +} + +static struct tcf_block *atm_tc_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow = (struct atm_flow_data *)cl; + + pr_debug("atm_tc_find_tcf(sch %p,[qdisc %p],flow %p)\n", sch, p, flow); + return flow ? flow->block : p->link.block; +} + +/* --------------------------- Qdisc operations ---------------------------- */ + +static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow; + struct tcf_result res; + int result; + int ret = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + + pr_debug("atm_tc_enqueue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p); + result = TC_ACT_OK; /* be nice to gcc */ + flow = NULL; + if (TC_H_MAJ(skb->priority) != sch->handle || + !(flow = (struct atm_flow_data *)atm_tc_find(sch, skb->priority))) { + struct tcf_proto *fl; + + list_for_each_entry(flow, &p->flows, list) { + fl = rcu_dereference_bh(flow->filter_list); + if (fl) { + result = tcf_classify(skb, fl, &res, true); + if (result < 0) + continue; + flow = (struct atm_flow_data *)res.class; + if (!flow) + flow = lookup_flow(sch, res.classid); + goto done; + } + } + flow = NULL; +done: + ; + } + if (!flow) { + flow = &p->link; + } else { + if (flow->vcc) + ATM_SKB(skb)->atm_options = flow->vcc->atm_options; + /*@@@ looks good ... but it's not supposed to work :-) */ +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + __qdisc_drop(skb, to_free); + return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + case TC_ACT_SHOT: + __qdisc_drop(skb, to_free); + goto drop; + case TC_ACT_RECLASSIFY: + if (flow->excess) + flow = flow->excess; + else + ATM_SKB(skb)->atm_options |= ATM_ATMOPT_CLP; + break; + } +#endif + } + + ret = qdisc_enqueue(skb, flow->q, to_free); + if (ret != NET_XMIT_SUCCESS) { +drop: __maybe_unused + if (net_xmit_drop_count(ret)) { + qdisc_qstats_drop(sch); + if (flow) + flow->qstats.drops++; + } + return ret; + } + /* + * Okay, this may seem weird. We pretend we've dropped the packet if + * it goes via ATM. The reason for this is that the outer qdisc + * expects to be able to q->dequeue the packet later on if we return + * success at this place. Also, sch->q.qdisc needs to reflect whether + * there is a packet egligible for dequeuing or not. Note that the + * statistics of the outer qdisc are necessarily wrong because of all + * this. There's currently no correct solution for this. + */ + if (flow == &p->link) { + sch->q.qlen++; + return NET_XMIT_SUCCESS; + } + tasklet_schedule(&p->task); + return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; +} + +/* + * Dequeue packets and send them over ATM. Note that we quite deliberately + * avoid checking net_device's flow control here, simply because sch_atm + * uses its own channels, which have nothing to do with any CLIP/LANE/or + * non-ATM interfaces. + */ + +static void sch_atm_dequeue(unsigned long data) +{ + struct Qdisc *sch = (struct Qdisc *)data; + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow; + struct sk_buff *skb; + + pr_debug("sch_atm_dequeue(sch %p,[qdisc %p])\n", sch, p); + list_for_each_entry(flow, &p->flows, list) { + if (flow == &p->link) + continue; + /* + * If traffic is properly shaped, this won't generate nasty + * little bursts. Otherwise, it may ... (but that's okay) + */ + while ((skb = flow->q->ops->peek(flow->q))) { + if (!atm_may_send(flow->vcc, skb->truesize)) + break; + + skb = qdisc_dequeue_peeked(flow->q); + if (unlikely(!skb)) + break; + + qdisc_bstats_update(sch, skb); + bstats_update(&flow->bstats, skb); + pr_debug("atm_tc_dequeue: sending on class %p\n", flow); + /* remove any LL header somebody else has attached */ + skb_pull(skb, skb_network_offset(skb)); + if (skb_headroom(skb) < flow->hdr_len) { + struct sk_buff *new; + + new = skb_realloc_headroom(skb, flow->hdr_len); + dev_kfree_skb(skb); + if (!new) + continue; + skb = new; + } + pr_debug("sch_atm_dequeue: ip %p, data %p\n", + skb_network_header(skb), skb->data); + ATM_SKB(skb)->vcc = flow->vcc; + memcpy(skb_push(skb, flow->hdr_len), flow->hdr, + flow->hdr_len); + refcount_add(skb->truesize, + &sk_atm(flow->vcc)->sk_wmem_alloc); + /* atm.atm_options are already set by atm_tc_enqueue */ + flow->vcc->send(flow->vcc, skb); + } + } +} + +static struct sk_buff *atm_tc_dequeue(struct Qdisc *sch) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct sk_buff *skb; + + pr_debug("atm_tc_dequeue(sch %p,[qdisc %p])\n", sch, p); + tasklet_schedule(&p->task); + skb = qdisc_dequeue_peeked(p->link.q); + if (skb) + sch->q.qlen--; + return skb; +} + +static struct sk_buff *atm_tc_peek(struct Qdisc *sch) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + + pr_debug("atm_tc_peek(sch %p,[qdisc %p])\n", sch, p); + + return p->link.q->ops->peek(p->link.q); +} + +static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + int err; + + pr_debug("atm_tc_init(sch %p,[qdisc %p],opt %p)\n", sch, p, opt); + INIT_LIST_HEAD(&p->flows); + INIT_LIST_HEAD(&p->link.list); + list_add(&p->link.list, &p->flows); + p->link.q = qdisc_create_dflt(sch->dev_queue, + &pfifo_qdisc_ops, sch->handle, extack); + if (!p->link.q) + p->link.q = &noop_qdisc; + pr_debug("atm_tc_init: link (%p) qdisc %p\n", &p->link, p->link.q); + p->link.vcc = NULL; + p->link.sock = NULL; + p->link.common.classid = sch->handle; + p->link.ref = 1; + + err = tcf_block_get(&p->link.block, &p->link.filter_list, sch, + extack); + if (err) + return err; + + tasklet_init(&p->task, sch_atm_dequeue, (unsigned long)sch); + return 0; +} + +static void atm_tc_reset(struct Qdisc *sch) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow; + + pr_debug("atm_tc_reset(sch %p,[qdisc %p])\n", sch, p); + list_for_each_entry(flow, &p->flows, list) + qdisc_reset(flow->q); + sch->q.qlen = 0; +} + +static void atm_tc_destroy(struct Qdisc *sch) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow, *tmp; + + pr_debug("atm_tc_destroy(sch %p,[qdisc %p])\n", sch, p); + list_for_each_entry(flow, &p->flows, list) { + tcf_block_put(flow->block); + flow->block = NULL; + } + + list_for_each_entry_safe(flow, tmp, &p->flows, list) { + if (flow->ref > 1) + pr_err("atm_destroy: %p->ref = %d\n", flow, flow->ref); + atm_tc_put(sch, (unsigned long)flow); + } + tasklet_kill(&p->task); +} + +static int atm_tc_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct atm_qdisc_data *p = qdisc_priv(sch); + struct atm_flow_data *flow = (struct atm_flow_data *)cl; + struct nlattr *nest; + + pr_debug("atm_tc_dump_class(sch %p,[qdisc %p],flow %p,skb %p,tcm %p)\n", + sch, p, flow, skb, tcm); + if (list_empty(&flow->list)) + return -EINVAL; + tcm->tcm_handle = flow->common.classid; + tcm->tcm_info = flow->q->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + if (nla_put(skb, TCA_ATM_HDR, flow->hdr_len, flow->hdr)) + goto nla_put_failure; + if (flow->vcc) { + struct sockaddr_atmpvc pvc; + int state; + + memset(&pvc, 0, sizeof(pvc)); + pvc.sap_family = AF_ATMPVC; + pvc.sap_addr.itf = flow->vcc->dev ? flow->vcc->dev->number : -1; + pvc.sap_addr.vpi = flow->vcc->vpi; + pvc.sap_addr.vci = flow->vcc->vci; + if (nla_put(skb, TCA_ATM_ADDR, sizeof(pvc), &pvc)) + goto nla_put_failure; + state = ATM_VF2VS(flow->vcc->flags); + if (nla_put_u32(skb, TCA_ATM_STATE, state)) + goto nla_put_failure; + } + if (flow->excess) { + if (nla_put_u32(skb, TCA_ATM_EXCESS, flow->common.classid)) + goto nla_put_failure; + } else { + if (nla_put_u32(skb, TCA_ATM_EXCESS, 0)) + goto nla_put_failure; + } + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} +static int +atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg, + struct gnet_dump *d) +{ + struct atm_flow_data *flow = (struct atm_flow_data *)arg; + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &flow->bstats) < 0 || + gnet_stats_copy_queue(d, NULL, &flow->qstats, flow->q->q.qlen) < 0) + return -1; + + return 0; +} + +static int atm_tc_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + return 0; +} + +static const struct Qdisc_class_ops atm_class_ops = { + .graft = atm_tc_graft, + .leaf = atm_tc_leaf, + .find = atm_tc_find, + .change = atm_tc_change, + .delete = atm_tc_delete, + .walk = atm_tc_walk, + .tcf_block = atm_tc_tcf_block, + .bind_tcf = atm_tc_bind_filter, + .unbind_tcf = atm_tc_put, + .dump = atm_tc_dump_class, + .dump_stats = atm_tc_dump_class_stats, +}; + +static struct Qdisc_ops atm_qdisc_ops __read_mostly = { + .cl_ops = &atm_class_ops, + .id = "atm", + .priv_size = sizeof(struct atm_qdisc_data), + .enqueue = atm_tc_enqueue, + .dequeue = atm_tc_dequeue, + .peek = atm_tc_peek, + .init = atm_tc_init, + .reset = atm_tc_reset, + .destroy = atm_tc_destroy, + .dump = atm_tc_dump, + .owner = THIS_MODULE, +}; + +static int __init atm_init(void) +{ + return register_qdisc(&atm_qdisc_ops); +} + +static void __exit atm_exit(void) +{ + unregister_qdisc(&atm_qdisc_ops); +} + +module_init(atm_init) +module_exit(atm_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_blackhole.c b/net/sched/sch_blackhole.c new file mode 100644 index 000000000..9c4c2bb54 --- /dev/null +++ b/net/sched/sch_blackhole.c @@ -0,0 +1,45 @@ +/* + * net/sched/sch_blackhole.c Black hole queue + * + * 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. + * + * Authors: Thomas Graf <tgraf@suug.ch> + * + * Note: Quantum tunneling is not supported. + */ + +#include <linux/init.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> + +static int blackhole_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + qdisc_drop(skb, sch, to_free); + return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; +} + +static struct sk_buff *blackhole_dequeue(struct Qdisc *sch) +{ + return NULL; +} + +static struct Qdisc_ops blackhole_qdisc_ops __read_mostly = { + .id = "blackhole", + .priv_size = 0, + .enqueue = blackhole_enqueue, + .dequeue = blackhole_dequeue, + .peek = blackhole_dequeue, + .owner = THIS_MODULE, +}; + +static int __init blackhole_init(void) +{ + return register_qdisc(&blackhole_qdisc_ops); +} +device_initcall(blackhole_init) diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c new file mode 100644 index 000000000..18c207b85 --- /dev/null +++ b/net/sched/sch_cake.c @@ -0,0 +1,3058 @@ +// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause + +/* COMMON Applications Kept Enhanced (CAKE) discipline + * + * Copyright (C) 2014-2018 Jonathan Morton <chromatix99@gmail.com> + * Copyright (C) 2015-2018 Toke Høiland-Jørgensen <toke@toke.dk> + * Copyright (C) 2014-2018 Dave Täht <dave.taht@gmail.com> + * Copyright (C) 2015-2018 Sebastian Moeller <moeller0@gmx.de> + * (C) 2015-2018 Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk> + * Copyright (C) 2017-2018 Ryan Mounce <ryan@mounce.com.au> + * + * The CAKE Principles: + * (or, how to have your cake and eat it too) + * + * This is a combination of several shaping, AQM and FQ techniques into one + * easy-to-use package: + * + * - An overall bandwidth shaper, to move the bottleneck away from dumb CPE + * equipment and bloated MACs. This operates in deficit mode (as in sch_fq), + * eliminating the need for any sort of burst parameter (eg. token bucket + * depth). Burst support is limited to that necessary to overcome scheduling + * latency. + * + * - A Diffserv-aware priority queue, giving more priority to certain classes, + * up to a specified fraction of bandwidth. Above that bandwidth threshold, + * the priority is reduced to avoid starving other tins. + * + * - Each priority tin has a separate Flow Queue system, to isolate traffic + * flows from each other. This prevents a burst on one flow from increasing + * the delay to another. Flows are distributed to queues using a + * set-associative hash function. + * + * - Each queue is actively managed by Cobalt, which is a combination of the + * Codel and Blue AQM algorithms. This serves flows fairly, and signals + * congestion early via ECN (if available) and/or packet drops, to keep + * latency low. The codel parameters are auto-tuned based on the bandwidth + * setting, as is necessary at low bandwidths. + * + * The configuration parameters are kept deliberately simple for ease of use. + * Everything has sane defaults. Complete generality of configuration is *not* + * a goal. + * + * The priority queue operates according to a weighted DRR scheme, combined with + * a bandwidth tracker which reuses the shaper logic to detect which side of the + * bandwidth sharing threshold the tin is operating. This determines whether a + * priority-based weight (high) or a bandwidth-based weight (low) is used for + * that tin in the current pass. + * + * This qdisc was inspired by Eric Dumazet's fq_codel code, which he kindly + * granted us permission to leverage. + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/jiffies.h> +#include <linux/string.h> +#include <linux/in.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/skbuff.h> +#include <linux/jhash.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <linux/reciprocal_div.h> +#include <net/netlink.h> +#include <linux/if_vlan.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/tcp.h> +#include <net/flow_dissector.h> + +#if IS_ENABLED(CONFIG_NF_CONNTRACK) +#include <net/netfilter/nf_conntrack_core.h> +#endif + +#define CAKE_SET_WAYS (8) +#define CAKE_MAX_TINS (8) +#define CAKE_QUEUES (1024) +#define CAKE_FLOW_MASK 63 +#define CAKE_FLOW_NAT_FLAG 64 + +/* struct cobalt_params - contains codel and blue parameters + * @interval: codel initial drop rate + * @target: maximum persistent sojourn time & blue update rate + * @mtu_time: serialisation delay of maximum-size packet + * @p_inc: increment of blue drop probability (0.32 fxp) + * @p_dec: decrement of blue drop probability (0.32 fxp) + */ +struct cobalt_params { + u64 interval; + u64 target; + u64 mtu_time; + u32 p_inc; + u32 p_dec; +}; + +/* struct cobalt_vars - contains codel and blue variables + * @count: codel dropping frequency + * @rec_inv_sqrt: reciprocal value of sqrt(count) >> 1 + * @drop_next: time to drop next packet, or when we dropped last + * @blue_timer: Blue time to next drop + * @p_drop: BLUE drop probability (0.32 fxp) + * @dropping: set if in dropping state + * @ecn_marked: set if marked + */ +struct cobalt_vars { + u32 count; + u32 rec_inv_sqrt; + ktime_t drop_next; + ktime_t blue_timer; + u32 p_drop; + bool dropping; + bool ecn_marked; +}; + +enum { + CAKE_SET_NONE = 0, + CAKE_SET_SPARSE, + CAKE_SET_SPARSE_WAIT, /* counted in SPARSE, actually in BULK */ + CAKE_SET_BULK, + CAKE_SET_DECAYING +}; + +struct cake_flow { + /* this stuff is all needed per-flow at dequeue time */ + struct sk_buff *head; + struct sk_buff *tail; + struct list_head flowchain; + s32 deficit; + u32 dropped; + struct cobalt_vars cvars; + u16 srchost; /* index into cake_host table */ + u16 dsthost; + u8 set; +}; /* please try to keep this structure <= 64 bytes */ + +struct cake_host { + u32 srchost_tag; + u32 dsthost_tag; + u16 srchost_refcnt; + u16 dsthost_refcnt; +}; + +struct cake_heap_entry { + u16 t:3, b:10; +}; + +struct cake_tin_data { + struct cake_flow flows[CAKE_QUEUES]; + u32 backlogs[CAKE_QUEUES]; + u32 tags[CAKE_QUEUES]; /* for set association */ + u16 overflow_idx[CAKE_QUEUES]; + struct cake_host hosts[CAKE_QUEUES]; /* for triple isolation */ + u16 flow_quantum; + + struct cobalt_params cparams; + u32 drop_overlimit; + u16 bulk_flow_count; + u16 sparse_flow_count; + u16 decaying_flow_count; + u16 unresponsive_flow_count; + + u32 max_skblen; + + struct list_head new_flows; + struct list_head old_flows; + struct list_head decaying_flows; + + /* time_next = time_this + ((len * rate_ns) >> rate_shft) */ + ktime_t time_next_packet; + u64 tin_rate_ns; + u64 tin_rate_bps; + u16 tin_rate_shft; + + u16 tin_quantum_prio; + u16 tin_quantum_band; + s32 tin_deficit; + u32 tin_backlog; + u32 tin_dropped; + u32 tin_ecn_mark; + + u32 packets; + u64 bytes; + + u32 ack_drops; + + /* moving averages */ + u64 avge_delay; + u64 peak_delay; + u64 base_delay; + + /* hash function stats */ + u32 way_directs; + u32 way_hits; + u32 way_misses; + u32 way_collisions; +}; /* number of tins is small, so size of this struct doesn't matter much */ + +struct cake_sched_data { + struct tcf_proto __rcu *filter_list; /* optional external classifier */ + struct tcf_block *block; + struct cake_tin_data *tins; + + struct cake_heap_entry overflow_heap[CAKE_QUEUES * CAKE_MAX_TINS]; + u16 overflow_timeout; + + u16 tin_cnt; + u8 tin_mode; + u8 flow_mode; + u8 ack_filter; + u8 atm_mode; + + /* time_next = time_this + ((len * rate_ns) >> rate_shft) */ + u16 rate_shft; + ktime_t time_next_packet; + ktime_t failsafe_next_packet; + u64 rate_ns; + u64 rate_bps; + u16 rate_flags; + s16 rate_overhead; + u16 rate_mpu; + u64 interval; + u64 target; + + /* resource tracking */ + u32 buffer_used; + u32 buffer_max_used; + u32 buffer_limit; + u32 buffer_config_limit; + + /* indices for dequeue */ + u16 cur_tin; + u16 cur_flow; + + struct qdisc_watchdog watchdog; + const u8 *tin_index; + const u8 *tin_order; + + /* bandwidth capacity estimate */ + ktime_t last_packet_time; + ktime_t avg_window_begin; + u64 avg_packet_interval; + u64 avg_window_bytes; + u64 avg_peak_bandwidth; + ktime_t last_reconfig_time; + + /* packet length stats */ + u32 avg_netoff; + u16 max_netlen; + u16 max_adjlen; + u16 min_netlen; + u16 min_adjlen; +}; + +enum { + CAKE_FLAG_OVERHEAD = BIT(0), + CAKE_FLAG_AUTORATE_INGRESS = BIT(1), + CAKE_FLAG_INGRESS = BIT(2), + CAKE_FLAG_WASH = BIT(3), + CAKE_FLAG_SPLIT_GSO = BIT(4) +}; + +/* COBALT operates the Codel and BLUE algorithms in parallel, in order to + * obtain the best features of each. Codel is excellent on flows which + * respond to congestion signals in a TCP-like way. BLUE is more effective on + * unresponsive flows. + */ + +struct cobalt_skb_cb { + ktime_t enqueue_time; + u32 adjusted_len; +}; + +static u64 us_to_ns(u64 us) +{ + return us * NSEC_PER_USEC; +} + +static struct cobalt_skb_cb *get_cobalt_cb(const struct sk_buff *skb) +{ + qdisc_cb_private_validate(skb, sizeof(struct cobalt_skb_cb)); + return (struct cobalt_skb_cb *)qdisc_skb_cb(skb)->data; +} + +static ktime_t cobalt_get_enqueue_time(const struct sk_buff *skb) +{ + return get_cobalt_cb(skb)->enqueue_time; +} + +static void cobalt_set_enqueue_time(struct sk_buff *skb, + ktime_t now) +{ + get_cobalt_cb(skb)->enqueue_time = now; +} + +static u16 quantum_div[CAKE_QUEUES + 1] = {0}; + +/* Diffserv lookup tables */ + +static const u8 precedence[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 1, 1, 1, 1, 1, 1, + 2, 2, 2, 2, 2, 2, 2, 2, + 3, 3, 3, 3, 3, 3, 3, 3, + 4, 4, 4, 4, 4, 4, 4, 4, + 5, 5, 5, 5, 5, 5, 5, 5, + 6, 6, 6, 6, 6, 6, 6, 6, + 7, 7, 7, 7, 7, 7, 7, 7, +}; + +static const u8 diffserv8[] = { + 2, 5, 1, 2, 4, 2, 2, 2, + 0, 2, 1, 2, 1, 2, 1, 2, + 5, 2, 4, 2, 4, 2, 4, 2, + 3, 2, 3, 2, 3, 2, 3, 2, + 6, 2, 3, 2, 3, 2, 3, 2, + 6, 2, 2, 2, 6, 2, 6, 2, + 7, 2, 2, 2, 2, 2, 2, 2, + 7, 2, 2, 2, 2, 2, 2, 2, +}; + +static const u8 diffserv4[] = { + 0, 2, 0, 0, 2, 0, 0, 0, + 1, 0, 0, 0, 0, 0, 0, 0, + 2, 0, 2, 0, 2, 0, 2, 0, + 2, 0, 2, 0, 2, 0, 2, 0, + 3, 0, 2, 0, 2, 0, 2, 0, + 3, 0, 0, 0, 3, 0, 3, 0, + 3, 0, 0, 0, 0, 0, 0, 0, + 3, 0, 0, 0, 0, 0, 0, 0, +}; + +static const u8 diffserv3[] = { + 0, 0, 0, 0, 2, 0, 0, 0, + 1, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 2, 0, 2, 0, + 2, 0, 0, 0, 0, 0, 0, 0, + 2, 0, 0, 0, 0, 0, 0, 0, +}; + +static const u8 besteffort[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, +}; + +/* tin priority order for stats dumping */ + +static const u8 normal_order[] = {0, 1, 2, 3, 4, 5, 6, 7}; +static const u8 bulk_order[] = {1, 0, 2, 3}; + +#define REC_INV_SQRT_CACHE (16) +static u32 cobalt_rec_inv_sqrt_cache[REC_INV_SQRT_CACHE] = {0}; + +/* http://en.wikipedia.org/wiki/Methods_of_computing_square_roots + * new_invsqrt = (invsqrt / 2) * (3 - count * invsqrt^2) + * + * Here, invsqrt is a fixed point number (< 1.0), 32bit mantissa, aka Q0.32 + */ + +static void cobalt_newton_step(struct cobalt_vars *vars) +{ + u32 invsqrt, invsqrt2; + u64 val; + + invsqrt = vars->rec_inv_sqrt; + invsqrt2 = ((u64)invsqrt * invsqrt) >> 32; + val = (3LL << 32) - ((u64)vars->count * invsqrt2); + + val >>= 2; /* avoid overflow in following multiply */ + val = (val * invsqrt) >> (32 - 2 + 1); + + vars->rec_inv_sqrt = val; +} + +static void cobalt_invsqrt(struct cobalt_vars *vars) +{ + if (vars->count < REC_INV_SQRT_CACHE) + vars->rec_inv_sqrt = cobalt_rec_inv_sqrt_cache[vars->count]; + else + cobalt_newton_step(vars); +} + +/* There is a big difference in timing between the accurate values placed in + * the cache and the approximations given by a single Newton step for small + * count values, particularly when stepping from count 1 to 2 or vice versa. + * Above 16, a single Newton step gives sufficient accuracy in either + * direction, given the precision stored. + * + * The magnitude of the error when stepping up to count 2 is such as to give + * the value that *should* have been produced at count 4. + */ + +static void cobalt_cache_init(void) +{ + struct cobalt_vars v; + + memset(&v, 0, sizeof(v)); + v.rec_inv_sqrt = ~0U; + cobalt_rec_inv_sqrt_cache[0] = v.rec_inv_sqrt; + + for (v.count = 1; v.count < REC_INV_SQRT_CACHE; v.count++) { + cobalt_newton_step(&v); + cobalt_newton_step(&v); + cobalt_newton_step(&v); + cobalt_newton_step(&v); + + cobalt_rec_inv_sqrt_cache[v.count] = v.rec_inv_sqrt; + } +} + +static void cobalt_vars_init(struct cobalt_vars *vars) +{ + memset(vars, 0, sizeof(*vars)); + + if (!cobalt_rec_inv_sqrt_cache[0]) { + cobalt_cache_init(); + cobalt_rec_inv_sqrt_cache[0] = ~0; + } +} + +/* CoDel control_law is t + interval/sqrt(count) + * We maintain in rec_inv_sqrt the reciprocal value of sqrt(count) to avoid + * both sqrt() and divide operation. + */ +static ktime_t cobalt_control(ktime_t t, + u64 interval, + u32 rec_inv_sqrt) +{ + return ktime_add_ns(t, reciprocal_scale(interval, + rec_inv_sqrt)); +} + +/* Call this when a packet had to be dropped due to queue overflow. Returns + * true if the BLUE state was quiescent before but active after this call. + */ +static bool cobalt_queue_full(struct cobalt_vars *vars, + struct cobalt_params *p, + ktime_t now) +{ + bool up = false; + + if (ktime_to_ns(ktime_sub(now, vars->blue_timer)) > p->target) { + up = !vars->p_drop; + vars->p_drop += p->p_inc; + if (vars->p_drop < p->p_inc) + vars->p_drop = ~0; + vars->blue_timer = now; + } + vars->dropping = true; + vars->drop_next = now; + if (!vars->count) + vars->count = 1; + + return up; +} + +/* Call this when the queue was serviced but turned out to be empty. Returns + * true if the BLUE state was active before but quiescent after this call. + */ +static bool cobalt_queue_empty(struct cobalt_vars *vars, + struct cobalt_params *p, + ktime_t now) +{ + bool down = false; + + if (vars->p_drop && + ktime_to_ns(ktime_sub(now, vars->blue_timer)) > p->target) { + if (vars->p_drop < p->p_dec) + vars->p_drop = 0; + else + vars->p_drop -= p->p_dec; + vars->blue_timer = now; + down = !vars->p_drop; + } + vars->dropping = false; + + if (vars->count && ktime_to_ns(ktime_sub(now, vars->drop_next)) >= 0) { + vars->count--; + cobalt_invsqrt(vars); + vars->drop_next = cobalt_control(vars->drop_next, + p->interval, + vars->rec_inv_sqrt); + } + + return down; +} + +/* Call this with a freshly dequeued packet for possible congestion marking. + * Returns true as an instruction to drop the packet, false for delivery. + */ +static bool cobalt_should_drop(struct cobalt_vars *vars, + struct cobalt_params *p, + ktime_t now, + struct sk_buff *skb, + u32 bulk_flows) +{ + bool next_due, over_target, drop = false; + ktime_t schedule; + u64 sojourn; + +/* The 'schedule' variable records, in its sign, whether 'now' is before or + * after 'drop_next'. This allows 'drop_next' to be updated before the next + * scheduling decision is actually branched, without destroying that + * information. Similarly, the first 'schedule' value calculated is preserved + * in the boolean 'next_due'. + * + * As for 'drop_next', we take advantage of the fact that 'interval' is both + * the delay between first exceeding 'target' and the first signalling event, + * *and* the scaling factor for the signalling frequency. It's therefore very + * natural to use a single mechanism for both purposes, and eliminates a + * significant amount of reference Codel's spaghetti code. To help with this, + * both the '0' and '1' entries in the invsqrt cache are 0xFFFFFFFF, as close + * as possible to 1.0 in fixed-point. + */ + + sojourn = ktime_to_ns(ktime_sub(now, cobalt_get_enqueue_time(skb))); + schedule = ktime_sub(now, vars->drop_next); + over_target = sojourn > p->target && + sojourn > p->mtu_time * bulk_flows * 2 && + sojourn > p->mtu_time * 4; + next_due = vars->count && ktime_to_ns(schedule) >= 0; + + vars->ecn_marked = false; + + if (over_target) { + if (!vars->dropping) { + vars->dropping = true; + vars->drop_next = cobalt_control(now, + p->interval, + vars->rec_inv_sqrt); + } + if (!vars->count) + vars->count = 1; + } else if (vars->dropping) { + vars->dropping = false; + } + + if (next_due && vars->dropping) { + /* Use ECN mark if possible, otherwise drop */ + drop = !(vars->ecn_marked = INET_ECN_set_ce(skb)); + + vars->count++; + if (!vars->count) + vars->count--; + cobalt_invsqrt(vars); + vars->drop_next = cobalt_control(vars->drop_next, + p->interval, + vars->rec_inv_sqrt); + schedule = ktime_sub(now, vars->drop_next); + } else { + while (next_due) { + vars->count--; + cobalt_invsqrt(vars); + vars->drop_next = cobalt_control(vars->drop_next, + p->interval, + vars->rec_inv_sqrt); + schedule = ktime_sub(now, vars->drop_next); + next_due = vars->count && ktime_to_ns(schedule) >= 0; + } + } + + /* Simple BLUE implementation. Lack of ECN is deliberate. */ + if (vars->p_drop) + drop |= (prandom_u32() < vars->p_drop); + + /* Overload the drop_next field as an activity timeout */ + if (!vars->count) + vars->drop_next = ktime_add_ns(now, p->interval); + else if (ktime_to_ns(schedule) > 0 && !drop) + vars->drop_next = now; + + return drop; +} + +static void cake_update_flowkeys(struct flow_keys *keys, + const struct sk_buff *skb) +{ +#if IS_ENABLED(CONFIG_NF_CONNTRACK) + struct nf_conntrack_tuple tuple = {}; + bool rev = !skb->_nfct; + + if (skb_protocol(skb, true) != htons(ETH_P_IP)) + return; + + if (!nf_ct_get_tuple_skb(&tuple, skb)) + return; + + keys->addrs.v4addrs.src = rev ? tuple.dst.u3.ip : tuple.src.u3.ip; + keys->addrs.v4addrs.dst = rev ? tuple.src.u3.ip : tuple.dst.u3.ip; + + if (keys->ports.ports) { + keys->ports.src = rev ? tuple.dst.u.all : tuple.src.u.all; + keys->ports.dst = rev ? tuple.src.u.all : tuple.dst.u.all; + } +#endif +} + +/* Cake has several subtle multiple bit settings. In these cases you + * would be matching triple isolate mode as well. + */ + +static bool cake_dsrc(int flow_mode) +{ + return (flow_mode & CAKE_FLOW_DUAL_SRC) == CAKE_FLOW_DUAL_SRC; +} + +static bool cake_ddst(int flow_mode) +{ + return (flow_mode & CAKE_FLOW_DUAL_DST) == CAKE_FLOW_DUAL_DST; +} + +static u32 cake_hash(struct cake_tin_data *q, const struct sk_buff *skb, + int flow_mode, u16 flow_override, u16 host_override) +{ + u32 flow_hash = 0, srchost_hash = 0, dsthost_hash = 0; + u16 reduced_hash, srchost_idx, dsthost_idx; + struct flow_keys keys, host_keys; + + if (unlikely(flow_mode == CAKE_FLOW_NONE)) + return 0; + + /* If both overrides are set we can skip packet dissection entirely */ + if ((flow_override || !(flow_mode & CAKE_FLOW_FLOWS)) && + (host_override || !(flow_mode & CAKE_FLOW_HOSTS))) + goto skip_hash; + + skb_flow_dissect_flow_keys(skb, &keys, + FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL); + + if (flow_mode & CAKE_FLOW_NAT_FLAG) + cake_update_flowkeys(&keys, skb); + + /* flow_hash_from_keys() sorts the addresses by value, so we have + * to preserve their order in a separate data structure to treat + * src and dst host addresses as independently selectable. + */ + host_keys = keys; + host_keys.ports.ports = 0; + host_keys.basic.ip_proto = 0; + host_keys.keyid.keyid = 0; + host_keys.tags.flow_label = 0; + + switch (host_keys.control.addr_type) { + case FLOW_DISSECTOR_KEY_IPV4_ADDRS: + host_keys.addrs.v4addrs.src = 0; + dsthost_hash = flow_hash_from_keys(&host_keys); + host_keys.addrs.v4addrs.src = keys.addrs.v4addrs.src; + host_keys.addrs.v4addrs.dst = 0; + srchost_hash = flow_hash_from_keys(&host_keys); + break; + + case FLOW_DISSECTOR_KEY_IPV6_ADDRS: + memset(&host_keys.addrs.v6addrs.src, 0, + sizeof(host_keys.addrs.v6addrs.src)); + dsthost_hash = flow_hash_from_keys(&host_keys); + host_keys.addrs.v6addrs.src = keys.addrs.v6addrs.src; + memset(&host_keys.addrs.v6addrs.dst, 0, + sizeof(host_keys.addrs.v6addrs.dst)); + srchost_hash = flow_hash_from_keys(&host_keys); + break; + + default: + dsthost_hash = 0; + srchost_hash = 0; + } + + /* This *must* be after the above switch, since as a + * side-effect it sorts the src and dst addresses. + */ + if (flow_mode & CAKE_FLOW_FLOWS) + flow_hash = flow_hash_from_keys(&keys); + +skip_hash: + if (flow_override) + flow_hash = flow_override - 1; + if (host_override) { + dsthost_hash = host_override - 1; + srchost_hash = host_override - 1; + } + + if (!(flow_mode & CAKE_FLOW_FLOWS)) { + if (flow_mode & CAKE_FLOW_SRC_IP) + flow_hash ^= srchost_hash; + + if (flow_mode & CAKE_FLOW_DST_IP) + flow_hash ^= dsthost_hash; + } + + reduced_hash = flow_hash % CAKE_QUEUES; + + /* set-associative hashing */ + /* fast path if no hash collision (direct lookup succeeds) */ + if (likely(q->tags[reduced_hash] == flow_hash && + q->flows[reduced_hash].set)) { + q->way_directs++; + } else { + u32 inner_hash = reduced_hash % CAKE_SET_WAYS; + u32 outer_hash = reduced_hash - inner_hash; + bool allocate_src = false; + bool allocate_dst = false; + u32 i, k; + + /* check if any active queue in the set is reserved for + * this flow. + */ + for (i = 0, k = inner_hash; i < CAKE_SET_WAYS; + i++, k = (k + 1) % CAKE_SET_WAYS) { + if (q->tags[outer_hash + k] == flow_hash) { + if (i) + q->way_hits++; + + if (!q->flows[outer_hash + k].set) { + /* need to increment host refcnts */ + allocate_src = cake_dsrc(flow_mode); + allocate_dst = cake_ddst(flow_mode); + } + + goto found; + } + } + + /* no queue is reserved for this flow, look for an + * empty one. + */ + for (i = 0; i < CAKE_SET_WAYS; + i++, k = (k + 1) % CAKE_SET_WAYS) { + if (!q->flows[outer_hash + k].set) { + q->way_misses++; + allocate_src = cake_dsrc(flow_mode); + allocate_dst = cake_ddst(flow_mode); + goto found; + } + } + + /* With no empty queues, default to the original + * queue, accept the collision, update the host tags. + */ + q->way_collisions++; + q->hosts[q->flows[reduced_hash].srchost].srchost_refcnt--; + q->hosts[q->flows[reduced_hash].dsthost].dsthost_refcnt--; + allocate_src = cake_dsrc(flow_mode); + allocate_dst = cake_ddst(flow_mode); +found: + /* reserve queue for future packets in same flow */ + reduced_hash = outer_hash + k; + q->tags[reduced_hash] = flow_hash; + + if (allocate_src) { + srchost_idx = srchost_hash % CAKE_QUEUES; + inner_hash = srchost_idx % CAKE_SET_WAYS; + outer_hash = srchost_idx - inner_hash; + for (i = 0, k = inner_hash; i < CAKE_SET_WAYS; + i++, k = (k + 1) % CAKE_SET_WAYS) { + if (q->hosts[outer_hash + k].srchost_tag == + srchost_hash) + goto found_src; + } + for (i = 0; i < CAKE_SET_WAYS; + i++, k = (k + 1) % CAKE_SET_WAYS) { + if (!q->hosts[outer_hash + k].srchost_refcnt) + break; + } + q->hosts[outer_hash + k].srchost_tag = srchost_hash; +found_src: + srchost_idx = outer_hash + k; + q->hosts[srchost_idx].srchost_refcnt++; + q->flows[reduced_hash].srchost = srchost_idx; + } + + if (allocate_dst) { + dsthost_idx = dsthost_hash % CAKE_QUEUES; + inner_hash = dsthost_idx % CAKE_SET_WAYS; + outer_hash = dsthost_idx - inner_hash; + for (i = 0, k = inner_hash; i < CAKE_SET_WAYS; + i++, k = (k + 1) % CAKE_SET_WAYS) { + if (q->hosts[outer_hash + k].dsthost_tag == + dsthost_hash) + goto found_dst; + } + for (i = 0; i < CAKE_SET_WAYS; + i++, k = (k + 1) % CAKE_SET_WAYS) { + if (!q->hosts[outer_hash + k].dsthost_refcnt) + break; + } + q->hosts[outer_hash + k].dsthost_tag = dsthost_hash; +found_dst: + dsthost_idx = outer_hash + k; + q->hosts[dsthost_idx].dsthost_refcnt++; + q->flows[reduced_hash].dsthost = dsthost_idx; + } + } + + return reduced_hash; +} + +/* helper functions : might be changed when/if skb use a standard list_head */ +/* remove one skb from head of slot queue */ + +static struct sk_buff *dequeue_head(struct cake_flow *flow) +{ + struct sk_buff *skb = flow->head; + + if (skb) { + flow->head = skb->next; + skb->next = NULL; + } + + return skb; +} + +/* add skb to flow queue (tail add) */ + +static void flow_queue_add(struct cake_flow *flow, struct sk_buff *skb) +{ + if (!flow->head) + flow->head = skb; + else + flow->tail->next = skb; + flow->tail = skb; + skb->next = NULL; +} + +static struct iphdr *cake_get_iphdr(const struct sk_buff *skb, + struct ipv6hdr *buf) +{ + unsigned int offset = skb_network_offset(skb); + struct iphdr *iph; + + iph = skb_header_pointer(skb, offset, sizeof(struct iphdr), buf); + + if (!iph) + return NULL; + + if (iph->version == 4 && iph->protocol == IPPROTO_IPV6) + return skb_header_pointer(skb, offset + iph->ihl * 4, + sizeof(struct ipv6hdr), buf); + + else if (iph->version == 4) + return iph; + + else if (iph->version == 6) + return skb_header_pointer(skb, offset, sizeof(struct ipv6hdr), + buf); + + return NULL; +} + +static struct tcphdr *cake_get_tcphdr(const struct sk_buff *skb, + void *buf, unsigned int bufsize) +{ + unsigned int offset = skb_network_offset(skb); + const struct ipv6hdr *ipv6h; + const struct tcphdr *tcph; + const struct iphdr *iph; + struct ipv6hdr _ipv6h; + struct tcphdr _tcph; + + ipv6h = skb_header_pointer(skb, offset, sizeof(_ipv6h), &_ipv6h); + + if (!ipv6h) + return NULL; + + if (ipv6h->version == 4) { + iph = (struct iphdr *)ipv6h; + offset += iph->ihl * 4; + + /* special-case 6in4 tunnelling, as that is a common way to get + * v6 connectivity in the home + */ + if (iph->protocol == IPPROTO_IPV6) { + ipv6h = skb_header_pointer(skb, offset, + sizeof(_ipv6h), &_ipv6h); + + if (!ipv6h || ipv6h->nexthdr != IPPROTO_TCP) + return NULL; + + offset += sizeof(struct ipv6hdr); + + } else if (iph->protocol != IPPROTO_TCP) { + return NULL; + } + + } else if (ipv6h->version == 6) { + if (ipv6h->nexthdr != IPPROTO_TCP) + return NULL; + + offset += sizeof(struct ipv6hdr); + } else { + return NULL; + } + + tcph = skb_header_pointer(skb, offset, sizeof(_tcph), &_tcph); + if (!tcph || tcph->doff < 5) + return NULL; + + return skb_header_pointer(skb, offset, + min(__tcp_hdrlen(tcph), bufsize), buf); +} + +static const void *cake_get_tcpopt(const struct tcphdr *tcph, + int code, int *oplen) +{ + /* inspired by tcp_parse_options in tcp_input.c */ + int length = __tcp_hdrlen(tcph) - sizeof(struct tcphdr); + const u8 *ptr = (const u8 *)(tcph + 1); + + while (length > 0) { + int opcode = *ptr++; + int opsize; + + if (opcode == TCPOPT_EOL) + break; + if (opcode == TCPOPT_NOP) { + length--; + continue; + } + if (length < 2) + break; + opsize = *ptr++; + if (opsize < 2 || opsize > length) + break; + + if (opcode == code) { + *oplen = opsize; + return ptr; + } + + ptr += opsize - 2; + length -= opsize; + } + + return NULL; +} + +/* Compare two SACK sequences. A sequence is considered greater if it SACKs more + * bytes than the other. In the case where both sequences ACKs bytes that the + * other doesn't, A is considered greater. DSACKs in A also makes A be + * considered greater. + * + * @return -1, 0 or 1 as normal compare functions + */ +static int cake_tcph_sack_compare(const struct tcphdr *tcph_a, + const struct tcphdr *tcph_b) +{ + const struct tcp_sack_block_wire *sack_a, *sack_b; + u32 ack_seq_a = ntohl(tcph_a->ack_seq); + u32 bytes_a = 0, bytes_b = 0; + int oplen_a, oplen_b; + bool first = true; + + sack_a = cake_get_tcpopt(tcph_a, TCPOPT_SACK, &oplen_a); + sack_b = cake_get_tcpopt(tcph_b, TCPOPT_SACK, &oplen_b); + + /* pointers point to option contents */ + oplen_a -= TCPOLEN_SACK_BASE; + oplen_b -= TCPOLEN_SACK_BASE; + + if (sack_a && oplen_a >= sizeof(*sack_a) && + (!sack_b || oplen_b < sizeof(*sack_b))) + return -1; + else if (sack_b && oplen_b >= sizeof(*sack_b) && + (!sack_a || oplen_a < sizeof(*sack_a))) + return 1; + else if ((!sack_a || oplen_a < sizeof(*sack_a)) && + (!sack_b || oplen_b < sizeof(*sack_b))) + return 0; + + while (oplen_a >= sizeof(*sack_a)) { + const struct tcp_sack_block_wire *sack_tmp = sack_b; + u32 start_a = get_unaligned_be32(&sack_a->start_seq); + u32 end_a = get_unaligned_be32(&sack_a->end_seq); + int oplen_tmp = oplen_b; + bool found = false; + + /* DSACK; always considered greater to prevent dropping */ + if (before(start_a, ack_seq_a)) + return -1; + + bytes_a += end_a - start_a; + + while (oplen_tmp >= sizeof(*sack_tmp)) { + u32 start_b = get_unaligned_be32(&sack_tmp->start_seq); + u32 end_b = get_unaligned_be32(&sack_tmp->end_seq); + + /* first time through we count the total size */ + if (first) + bytes_b += end_b - start_b; + + if (!after(start_b, start_a) && !before(end_b, end_a)) { + found = true; + if (!first) + break; + } + oplen_tmp -= sizeof(*sack_tmp); + sack_tmp++; + } + + if (!found) + return -1; + + oplen_a -= sizeof(*sack_a); + sack_a++; + first = false; + } + + /* If we made it this far, all ranges SACKed by A are covered by B, so + * either the SACKs are equal, or B SACKs more bytes. + */ + return bytes_b > bytes_a ? 1 : 0; +} + +static void cake_tcph_get_tstamp(const struct tcphdr *tcph, + u32 *tsval, u32 *tsecr) +{ + const u8 *ptr; + int opsize; + + ptr = cake_get_tcpopt(tcph, TCPOPT_TIMESTAMP, &opsize); + + if (ptr && opsize == TCPOLEN_TIMESTAMP) { + *tsval = get_unaligned_be32(ptr); + *tsecr = get_unaligned_be32(ptr + 4); + } +} + +static bool cake_tcph_may_drop(const struct tcphdr *tcph, + u32 tstamp_new, u32 tsecr_new) +{ + /* inspired by tcp_parse_options in tcp_input.c */ + int length = __tcp_hdrlen(tcph) - sizeof(struct tcphdr); + const u8 *ptr = (const u8 *)(tcph + 1); + u32 tstamp, tsecr; + + /* 3 reserved flags must be unset to avoid future breakage + * ACK must be set + * ECE/CWR are handled separately + * All other flags URG/PSH/RST/SYN/FIN must be unset + * 0x0FFF0000 = all TCP flags (confirm ACK=1, others zero) + * 0x00C00000 = CWR/ECE (handled separately) + * 0x0F3F0000 = 0x0FFF0000 & ~0x00C00000 + */ + if (((tcp_flag_word(tcph) & + cpu_to_be32(0x0F3F0000)) != TCP_FLAG_ACK)) + return false; + + while (length > 0) { + int opcode = *ptr++; + int opsize; + + if (opcode == TCPOPT_EOL) + break; + if (opcode == TCPOPT_NOP) { + length--; + continue; + } + if (length < 2) + break; + opsize = *ptr++; + if (opsize < 2 || opsize > length) + break; + + switch (opcode) { + case TCPOPT_MD5SIG: /* doesn't influence state */ + break; + + case TCPOPT_SACK: /* stricter checking performed later */ + if (opsize % 8 != 2) + return false; + break; + + case TCPOPT_TIMESTAMP: + /* only drop timestamps lower than new */ + if (opsize != TCPOLEN_TIMESTAMP) + return false; + tstamp = get_unaligned_be32(ptr); + tsecr = get_unaligned_be32(ptr + 4); + if (after(tstamp, tstamp_new) || + after(tsecr, tsecr_new)) + return false; + break; + + case TCPOPT_MSS: /* these should only be set on SYN */ + case TCPOPT_WINDOW: + case TCPOPT_SACK_PERM: + case TCPOPT_FASTOPEN: + case TCPOPT_EXP: + default: /* don't drop if any unknown options are present */ + return false; + } + + ptr += opsize - 2; + length -= opsize; + } + + return true; +} + +static struct sk_buff *cake_ack_filter(struct cake_sched_data *q, + struct cake_flow *flow) +{ + bool aggressive = q->ack_filter == CAKE_ACK_AGGRESSIVE; + struct sk_buff *elig_ack = NULL, *elig_ack_prev = NULL; + struct sk_buff *skb_check, *skb_prev = NULL; + const struct ipv6hdr *ipv6h, *ipv6h_check; + unsigned char _tcph[64], _tcph_check[64]; + const struct tcphdr *tcph, *tcph_check; + const struct iphdr *iph, *iph_check; + struct ipv6hdr _iph, _iph_check; + const struct sk_buff *skb; + int seglen, num_found = 0; + u32 tstamp = 0, tsecr = 0; + __be32 elig_flags = 0; + int sack_comp; + + /* no other possible ACKs to filter */ + if (flow->head == flow->tail) + return NULL; + + skb = flow->tail; + tcph = cake_get_tcphdr(skb, _tcph, sizeof(_tcph)); + iph = cake_get_iphdr(skb, &_iph); + if (!tcph) + return NULL; + + cake_tcph_get_tstamp(tcph, &tstamp, &tsecr); + + /* the 'triggering' packet need only have the ACK flag set. + * also check that SYN is not set, as there won't be any previous ACKs. + */ + if ((tcp_flag_word(tcph) & + (TCP_FLAG_ACK | TCP_FLAG_SYN)) != TCP_FLAG_ACK) + return NULL; + + /* the 'triggering' ACK is at the tail of the queue, we have already + * returned if it is the only packet in the flow. loop through the rest + * of the queue looking for pure ACKs with the same 5-tuple as the + * triggering one. + */ + for (skb_check = flow->head; + skb_check && skb_check != skb; + skb_prev = skb_check, skb_check = skb_check->next) { + iph_check = cake_get_iphdr(skb_check, &_iph_check); + tcph_check = cake_get_tcphdr(skb_check, &_tcph_check, + sizeof(_tcph_check)); + + /* only TCP packets with matching 5-tuple are eligible, and only + * drop safe headers + */ + if (!tcph_check || iph->version != iph_check->version || + tcph_check->source != tcph->source || + tcph_check->dest != tcph->dest) + continue; + + if (iph_check->version == 4) { + if (iph_check->saddr != iph->saddr || + iph_check->daddr != iph->daddr) + continue; + + seglen = ntohs(iph_check->tot_len) - + (4 * iph_check->ihl); + } else if (iph_check->version == 6) { + ipv6h = (struct ipv6hdr *)iph; + ipv6h_check = (struct ipv6hdr *)iph_check; + + if (ipv6_addr_cmp(&ipv6h_check->saddr, &ipv6h->saddr) || + ipv6_addr_cmp(&ipv6h_check->daddr, &ipv6h->daddr)) + continue; + + seglen = ntohs(ipv6h_check->payload_len); + } else { + WARN_ON(1); /* shouldn't happen */ + continue; + } + + /* If the ECE/CWR flags changed from the previous eligible + * packet in the same flow, we should no longer be dropping that + * previous packet as this would lose information. + */ + if (elig_ack && (tcp_flag_word(tcph_check) & + (TCP_FLAG_ECE | TCP_FLAG_CWR)) != elig_flags) { + elig_ack = NULL; + elig_ack_prev = NULL; + num_found--; + } + + /* Check TCP options and flags, don't drop ACKs with segment + * data, and don't drop ACKs with a higher cumulative ACK + * counter than the triggering packet. Check ACK seqno here to + * avoid parsing SACK options of packets we are going to exclude + * anyway. + */ + if (!cake_tcph_may_drop(tcph_check, tstamp, tsecr) || + (seglen - __tcp_hdrlen(tcph_check)) != 0 || + after(ntohl(tcph_check->ack_seq), ntohl(tcph->ack_seq))) + continue; + + /* Check SACK options. The triggering packet must SACK more data + * than the ACK under consideration, or SACK the same range but + * have a larger cumulative ACK counter. The latter is a + * pathological case, but is contained in the following check + * anyway, just to be safe. + */ + sack_comp = cake_tcph_sack_compare(tcph_check, tcph); + + if (sack_comp < 0 || + (ntohl(tcph_check->ack_seq) == ntohl(tcph->ack_seq) && + sack_comp == 0)) + continue; + + /* At this point we have found an eligible pure ACK to drop; if + * we are in aggressive mode, we are done. Otherwise, keep + * searching unless this is the second eligible ACK we + * found. + * + * Since we want to drop ACK closest to the head of the queue, + * save the first eligible ACK we find, even if we need to loop + * again. + */ + if (!elig_ack) { + elig_ack = skb_check; + elig_ack_prev = skb_prev; + elig_flags = (tcp_flag_word(tcph_check) + & (TCP_FLAG_ECE | TCP_FLAG_CWR)); + } + + if (num_found++ > 0) + goto found; + } + + /* We made it through the queue without finding two eligible ACKs . If + * we found a single eligible ACK we can drop it in aggressive mode if + * we can guarantee that this does not interfere with ECN flag + * information. We ensure this by dropping it only if the enqueued + * packet is consecutive with the eligible ACK, and their flags match. + */ + if (elig_ack && aggressive && elig_ack->next == skb && + (elig_flags == (tcp_flag_word(tcph) & + (TCP_FLAG_ECE | TCP_FLAG_CWR)))) + goto found; + + return NULL; + +found: + if (elig_ack_prev) + elig_ack_prev->next = elig_ack->next; + else + flow->head = elig_ack->next; + + elig_ack->next = NULL; + + return elig_ack; +} + +static u64 cake_ewma(u64 avg, u64 sample, u32 shift) +{ + avg -= avg >> shift; + avg += sample >> shift; + return avg; +} + +static u32 cake_calc_overhead(struct cake_sched_data *q, u32 len, u32 off) +{ + if (q->rate_flags & CAKE_FLAG_OVERHEAD) + len -= off; + + if (q->max_netlen < len) + q->max_netlen = len; + if (q->min_netlen > len) + q->min_netlen = len; + + len += q->rate_overhead; + + if (len < q->rate_mpu) + len = q->rate_mpu; + + if (q->atm_mode == CAKE_ATM_ATM) { + len += 47; + len /= 48; + len *= 53; + } else if (q->atm_mode == CAKE_ATM_PTM) { + /* Add one byte per 64 bytes or part thereof. + * This is conservative and easier to calculate than the + * precise value. + */ + len += (len + 63) / 64; + } + + if (q->max_adjlen < len) + q->max_adjlen = len; + if (q->min_adjlen > len) + q->min_adjlen = len; + + return len; +} + +static u32 cake_overhead(struct cake_sched_data *q, const struct sk_buff *skb) +{ + const struct skb_shared_info *shinfo = skb_shinfo(skb); + unsigned int hdr_len, last_len = 0; + u32 off = skb_network_offset(skb); + u32 len = qdisc_pkt_len(skb); + u16 segs = 1; + + q->avg_netoff = cake_ewma(q->avg_netoff, off << 16, 8); + + if (!shinfo->gso_size) + return cake_calc_overhead(q, len, off); + + /* borrowed from qdisc_pkt_len_init() */ + hdr_len = skb_transport_header(skb) - skb_mac_header(skb); + + /* + transport layer */ + if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | + SKB_GSO_TCPV6))) { + const struct tcphdr *th; + struct tcphdr _tcphdr; + + th = skb_header_pointer(skb, skb_transport_offset(skb), + sizeof(_tcphdr), &_tcphdr); + if (likely(th)) + hdr_len += __tcp_hdrlen(th); + } else { + struct udphdr _udphdr; + + if (skb_header_pointer(skb, skb_transport_offset(skb), + sizeof(_udphdr), &_udphdr)) + hdr_len += sizeof(struct udphdr); + } + + if (unlikely(shinfo->gso_type & SKB_GSO_DODGY)) + segs = DIV_ROUND_UP(skb->len - hdr_len, + shinfo->gso_size); + else + segs = shinfo->gso_segs; + + len = shinfo->gso_size + hdr_len; + last_len = skb->len - shinfo->gso_size * (segs - 1); + + return (cake_calc_overhead(q, len, off) * (segs - 1) + + cake_calc_overhead(q, last_len, off)); +} + +static void cake_heap_swap(struct cake_sched_data *q, u16 i, u16 j) +{ + struct cake_heap_entry ii = q->overflow_heap[i]; + struct cake_heap_entry jj = q->overflow_heap[j]; + + q->overflow_heap[i] = jj; + q->overflow_heap[j] = ii; + + q->tins[ii.t].overflow_idx[ii.b] = j; + q->tins[jj.t].overflow_idx[jj.b] = i; +} + +static u32 cake_heap_get_backlog(const struct cake_sched_data *q, u16 i) +{ + struct cake_heap_entry ii = q->overflow_heap[i]; + + return q->tins[ii.t].backlogs[ii.b]; +} + +static void cake_heapify(struct cake_sched_data *q, u16 i) +{ + static const u32 a = CAKE_MAX_TINS * CAKE_QUEUES; + u32 mb = cake_heap_get_backlog(q, i); + u32 m = i; + + while (m < a) { + u32 l = m + m + 1; + u32 r = l + 1; + + if (l < a) { + u32 lb = cake_heap_get_backlog(q, l); + + if (lb > mb) { + m = l; + mb = lb; + } + } + + if (r < a) { + u32 rb = cake_heap_get_backlog(q, r); + + if (rb > mb) { + m = r; + mb = rb; + } + } + + if (m != i) { + cake_heap_swap(q, i, m); + i = m; + } else { + break; + } + } +} + +static void cake_heapify_up(struct cake_sched_data *q, u16 i) +{ + while (i > 0 && i < CAKE_MAX_TINS * CAKE_QUEUES) { + u16 p = (i - 1) >> 1; + u32 ib = cake_heap_get_backlog(q, i); + u32 pb = cake_heap_get_backlog(q, p); + + if (ib > pb) { + cake_heap_swap(q, i, p); + i = p; + } else { + break; + } + } +} + +static int cake_advance_shaper(struct cake_sched_data *q, + struct cake_tin_data *b, + struct sk_buff *skb, + ktime_t now, bool drop) +{ + u32 len = get_cobalt_cb(skb)->adjusted_len; + + /* charge packet bandwidth to this tin + * and to the global shaper. + */ + if (q->rate_ns) { + u64 tin_dur = (len * b->tin_rate_ns) >> b->tin_rate_shft; + u64 global_dur = (len * q->rate_ns) >> q->rate_shft; + u64 failsafe_dur = global_dur + (global_dur >> 1); + + if (ktime_before(b->time_next_packet, now)) + b->time_next_packet = ktime_add_ns(b->time_next_packet, + tin_dur); + + else if (ktime_before(b->time_next_packet, + ktime_add_ns(now, tin_dur))) + b->time_next_packet = ktime_add_ns(now, tin_dur); + + q->time_next_packet = ktime_add_ns(q->time_next_packet, + global_dur); + if (!drop) + q->failsafe_next_packet = \ + ktime_add_ns(q->failsafe_next_packet, + failsafe_dur); + } + return len; +} + +static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) +{ + struct cake_sched_data *q = qdisc_priv(sch); + ktime_t now = ktime_get(); + u32 idx = 0, tin = 0, len; + struct cake_heap_entry qq; + struct cake_tin_data *b; + struct cake_flow *flow; + struct sk_buff *skb; + + if (!q->overflow_timeout) { + int i; + /* Build fresh max-heap */ + for (i = CAKE_MAX_TINS * CAKE_QUEUES / 2; i >= 0; i--) + cake_heapify(q, i); + } + q->overflow_timeout = 65535; + + /* select longest queue for pruning */ + qq = q->overflow_heap[0]; + tin = qq.t; + idx = qq.b; + + b = &q->tins[tin]; + flow = &b->flows[idx]; + skb = dequeue_head(flow); + if (unlikely(!skb)) { + /* heap has gone wrong, rebuild it next time */ + q->overflow_timeout = 0; + return idx + (tin << 16); + } + + if (cobalt_queue_full(&flow->cvars, &b->cparams, now)) + b->unresponsive_flow_count++; + + len = qdisc_pkt_len(skb); + q->buffer_used -= skb->truesize; + b->backlogs[idx] -= len; + b->tin_backlog -= len; + sch->qstats.backlog -= len; + qdisc_tree_reduce_backlog(sch, 1, len); + + flow->dropped++; + b->tin_dropped++; + sch->qstats.drops++; + + if (q->rate_flags & CAKE_FLAG_INGRESS) + cake_advance_shaper(q, b, skb, now, true); + + __qdisc_drop(skb, to_free); + sch->q.qlen--; + + cake_heapify(q, 0); + + return idx + (tin << 16); +} + +static u8 cake_handle_diffserv(struct sk_buff *skb, bool wash) +{ + const int offset = skb_network_offset(skb); + u16 *buf, buf_; + u8 dscp; + + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + buf = skb_header_pointer(skb, offset, sizeof(buf_), &buf_); + if (unlikely(!buf)) + return 0; + + /* ToS is in the second byte of iphdr */ + dscp = ipv4_get_dsfield((struct iphdr *)buf) >> 2; + + if (wash && dscp) { + const int wlen = offset + sizeof(struct iphdr); + + if (!pskb_may_pull(skb, wlen) || + skb_try_make_writable(skb, wlen)) + return 0; + + ipv4_change_dsfield(ip_hdr(skb), INET_ECN_MASK, 0); + } + + return dscp; + + case htons(ETH_P_IPV6): + buf = skb_header_pointer(skb, offset, sizeof(buf_), &buf_); + if (unlikely(!buf)) + return 0; + + /* Traffic class is in the first and second bytes of ipv6hdr */ + dscp = ipv6_get_dsfield((struct ipv6hdr *)buf) >> 2; + + if (wash && dscp) { + const int wlen = offset + sizeof(struct ipv6hdr); + + if (!pskb_may_pull(skb, wlen) || + skb_try_make_writable(skb, wlen)) + return 0; + + ipv6_change_dsfield(ipv6_hdr(skb), INET_ECN_MASK, 0); + } + + return dscp; + + case htons(ETH_P_ARP): + return 0x38; /* CS7 - Net Control */ + + default: + /* If there is no Diffserv field, treat as best-effort */ + return 0; + } +} + +static struct cake_tin_data *cake_select_tin(struct Qdisc *sch, + struct sk_buff *skb) +{ + struct cake_sched_data *q = qdisc_priv(sch); + u32 tin; + bool wash; + u8 dscp; + + /* Tin selection: Default to diffserv-based selection, allow overriding + * using firewall marks or skb->priority. Call DSCP parsing early if + * wash is enabled, otherwise defer to below to skip unneeded parsing. + */ + wash = !!(q->rate_flags & CAKE_FLAG_WASH); + if (wash) + dscp = cake_handle_diffserv(skb, wash); + + if (q->tin_mode == CAKE_DIFFSERV_BESTEFFORT) + tin = 0; + + else if (TC_H_MAJ(skb->priority) == sch->handle && + TC_H_MIN(skb->priority) > 0 && + TC_H_MIN(skb->priority) <= q->tin_cnt) + tin = q->tin_order[TC_H_MIN(skb->priority) - 1]; + + else { + if (!wash) + dscp = cake_handle_diffserv(skb, wash); + tin = q->tin_index[dscp]; + + if (unlikely(tin >= q->tin_cnt)) + tin = 0; + } + + return &q->tins[tin]; +} + +static u32 cake_classify(struct Qdisc *sch, struct cake_tin_data **t, + struct sk_buff *skb, int flow_mode, int *qerr) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct tcf_proto *filter; + struct tcf_result res; + u16 flow = 0, host = 0; + int result; + + filter = rcu_dereference_bh(q->filter_list); + if (!filter) + goto hash; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + result = tcf_classify(skb, filter, &res, false); + + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return 0; + } +#endif + if (TC_H_MIN(res.classid) <= CAKE_QUEUES) + flow = TC_H_MIN(res.classid); + if (TC_H_MAJ(res.classid) <= (CAKE_QUEUES << 16)) + host = TC_H_MAJ(res.classid) >> 16; + } +hash: + *t = cake_select_tin(sch, skb); + return cake_hash(*t, skb, flow_mode, flow, host) + 1; +} + +static void cake_reconfigure(struct Qdisc *sch); + +static s32 cake_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct cake_sched_data *q = qdisc_priv(sch); + int len = qdisc_pkt_len(skb); + int uninitialized_var(ret); + struct sk_buff *ack = NULL; + ktime_t now = ktime_get(); + struct cake_tin_data *b; + struct cake_flow *flow; + u32 idx; + + /* choose flow to insert into */ + idx = cake_classify(sch, &b, skb, q->flow_mode, &ret); + if (idx == 0) { + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; + } + idx--; + flow = &b->flows[idx]; + + /* ensure shaper state isn't stale */ + if (!b->tin_backlog) { + if (ktime_before(b->time_next_packet, now)) + b->time_next_packet = now; + + if (!sch->q.qlen) { + if (ktime_before(q->time_next_packet, now)) { + q->failsafe_next_packet = now; + q->time_next_packet = now; + } else if (ktime_after(q->time_next_packet, now) && + ktime_after(q->failsafe_next_packet, now)) { + u64 next = \ + min(ktime_to_ns(q->time_next_packet), + ktime_to_ns( + q->failsafe_next_packet)); + sch->qstats.overlimits++; + qdisc_watchdog_schedule_ns(&q->watchdog, next); + } + } + } + + if (unlikely(len > b->max_skblen)) + b->max_skblen = len; + + if (skb_is_gso(skb) && q->rate_flags & CAKE_FLAG_SPLIT_GSO) { + struct sk_buff *segs, *nskb; + netdev_features_t features = netif_skb_features(skb); + unsigned int slen = 0, numsegs = 0; + + segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); + if (IS_ERR_OR_NULL(segs)) + return qdisc_drop(skb, sch, to_free); + + while (segs) { + nskb = segs->next; + segs->next = NULL; + qdisc_skb_cb(segs)->pkt_len = segs->len; + cobalt_set_enqueue_time(segs, now); + get_cobalt_cb(segs)->adjusted_len = cake_overhead(q, + segs); + flow_queue_add(flow, segs); + + sch->q.qlen++; + numsegs++; + slen += segs->len; + q->buffer_used += segs->truesize; + b->packets++; + segs = nskb; + } + + /* stats */ + b->bytes += slen; + b->backlogs[idx] += slen; + b->tin_backlog += slen; + sch->qstats.backlog += slen; + q->avg_window_bytes += slen; + + qdisc_tree_reduce_backlog(sch, 1-numsegs, len-slen); + consume_skb(skb); + } else { + /* not splitting */ + cobalt_set_enqueue_time(skb, now); + get_cobalt_cb(skb)->adjusted_len = cake_overhead(q, skb); + flow_queue_add(flow, skb); + + if (q->ack_filter) + ack = cake_ack_filter(q, flow); + + if (ack) { + b->ack_drops++; + sch->qstats.drops++; + b->bytes += qdisc_pkt_len(ack); + len -= qdisc_pkt_len(ack); + q->buffer_used += skb->truesize - ack->truesize; + if (q->rate_flags & CAKE_FLAG_INGRESS) + cake_advance_shaper(q, b, ack, now, true); + + qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(ack)); + consume_skb(ack); + } else { + sch->q.qlen++; + q->buffer_used += skb->truesize; + } + + /* stats */ + b->packets++; + b->bytes += len; + b->backlogs[idx] += len; + b->tin_backlog += len; + sch->qstats.backlog += len; + q->avg_window_bytes += len; + } + + if (q->overflow_timeout) + cake_heapify_up(q, b->overflow_idx[idx]); + + /* incoming bandwidth capacity estimate */ + if (q->rate_flags & CAKE_FLAG_AUTORATE_INGRESS) { + u64 packet_interval = \ + ktime_to_ns(ktime_sub(now, q->last_packet_time)); + + if (packet_interval > NSEC_PER_SEC) + packet_interval = NSEC_PER_SEC; + + /* filter out short-term bursts, eg. wifi aggregation */ + q->avg_packet_interval = \ + cake_ewma(q->avg_packet_interval, + packet_interval, + (packet_interval > q->avg_packet_interval ? + 2 : 8)); + + q->last_packet_time = now; + + if (packet_interval > q->avg_packet_interval) { + u64 window_interval = \ + ktime_to_ns(ktime_sub(now, + q->avg_window_begin)); + u64 b = q->avg_window_bytes * (u64)NSEC_PER_SEC; + + b = div64_u64(b, window_interval); + q->avg_peak_bandwidth = + cake_ewma(q->avg_peak_bandwidth, b, + b > q->avg_peak_bandwidth ? 2 : 8); + q->avg_window_bytes = 0; + q->avg_window_begin = now; + + if (ktime_after(now, + ktime_add_ms(q->last_reconfig_time, + 250))) { + q->rate_bps = (q->avg_peak_bandwidth * 15) >> 4; + cake_reconfigure(sch); + } + } + } else { + q->avg_window_bytes = 0; + q->last_packet_time = now; + } + + /* flowchain */ + if (!flow->set || flow->set == CAKE_SET_DECAYING) { + struct cake_host *srchost = &b->hosts[flow->srchost]; + struct cake_host *dsthost = &b->hosts[flow->dsthost]; + u16 host_load = 1; + + if (!flow->set) { + list_add_tail(&flow->flowchain, &b->new_flows); + } else { + b->decaying_flow_count--; + list_move_tail(&flow->flowchain, &b->new_flows); + } + flow->set = CAKE_SET_SPARSE; + b->sparse_flow_count++; + + if (cake_dsrc(q->flow_mode)) + host_load = max(host_load, srchost->srchost_refcnt); + + if (cake_ddst(q->flow_mode)) + host_load = max(host_load, dsthost->dsthost_refcnt); + + flow->deficit = (b->flow_quantum * + quantum_div[host_load]) >> 16; + } else if (flow->set == CAKE_SET_SPARSE_WAIT) { + /* this flow was empty, accounted as a sparse flow, but actually + * in the bulk rotation. + */ + flow->set = CAKE_SET_BULK; + b->sparse_flow_count--; + b->bulk_flow_count++; + } + + if (q->buffer_used > q->buffer_max_used) + q->buffer_max_used = q->buffer_used; + + if (q->buffer_used > q->buffer_limit) { + u32 dropped = 0; + + while (q->buffer_used > q->buffer_limit) { + dropped++; + cake_drop(sch, to_free); + } + b->drop_overlimit += dropped; + } + return NET_XMIT_SUCCESS; +} + +static struct sk_buff *cake_dequeue_one(struct Qdisc *sch) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct cake_tin_data *b = &q->tins[q->cur_tin]; + struct cake_flow *flow = &b->flows[q->cur_flow]; + struct sk_buff *skb = NULL; + u32 len; + + if (flow->head) { + skb = dequeue_head(flow); + len = qdisc_pkt_len(skb); + b->backlogs[q->cur_flow] -= len; + b->tin_backlog -= len; + sch->qstats.backlog -= len; + q->buffer_used -= skb->truesize; + sch->q.qlen--; + + if (q->overflow_timeout) + cake_heapify(q, b->overflow_idx[q->cur_flow]); + } + return skb; +} + +/* Discard leftover packets from a tin no longer in use. */ +static void cake_clear_tin(struct Qdisc *sch, u16 tin) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + + q->cur_tin = tin; + for (q->cur_flow = 0; q->cur_flow < CAKE_QUEUES; q->cur_flow++) + while (!!(skb = cake_dequeue_one(sch))) + kfree_skb(skb); +} + +static struct sk_buff *cake_dequeue(struct Qdisc *sch) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct cake_tin_data *b = &q->tins[q->cur_tin]; + struct cake_host *srchost, *dsthost; + ktime_t now = ktime_get(); + struct cake_flow *flow; + struct list_head *head; + bool first_flow = true; + struct sk_buff *skb; + u16 host_load; + u64 delay; + u32 len; + +begin: + if (!sch->q.qlen) + return NULL; + + /* global hard shaper */ + if (ktime_after(q->time_next_packet, now) && + ktime_after(q->failsafe_next_packet, now)) { + u64 next = min(ktime_to_ns(q->time_next_packet), + ktime_to_ns(q->failsafe_next_packet)); + + sch->qstats.overlimits++; + qdisc_watchdog_schedule_ns(&q->watchdog, next); + return NULL; + } + + /* Choose a class to work on. */ + if (!q->rate_ns) { + /* In unlimited mode, can't rely on shaper timings, just balance + * with DRR + */ + bool wrapped = false, empty = true; + + while (b->tin_deficit < 0 || + !(b->sparse_flow_count + b->bulk_flow_count)) { + if (b->tin_deficit <= 0) + b->tin_deficit += b->tin_quantum_band; + if (b->sparse_flow_count + b->bulk_flow_count) + empty = false; + + q->cur_tin++; + b++; + if (q->cur_tin >= q->tin_cnt) { + q->cur_tin = 0; + b = q->tins; + + if (wrapped) { + /* It's possible for q->qlen to be + * nonzero when we actually have no + * packets anywhere. + */ + if (empty) + return NULL; + } else { + wrapped = true; + } + } + } + } else { + /* In shaped mode, choose: + * - Highest-priority tin with queue and meeting schedule, or + * - The earliest-scheduled tin with queue. + */ + ktime_t best_time = KTIME_MAX; + int tin, best_tin = 0; + + for (tin = 0; tin < q->tin_cnt; tin++) { + b = q->tins + tin; + if ((b->sparse_flow_count + b->bulk_flow_count) > 0) { + ktime_t time_to_pkt = \ + ktime_sub(b->time_next_packet, now); + + if (ktime_to_ns(time_to_pkt) <= 0 || + ktime_compare(time_to_pkt, + best_time) <= 0) { + best_time = time_to_pkt; + best_tin = tin; + } + } + } + + q->cur_tin = best_tin; + b = q->tins + best_tin; + + /* No point in going further if no packets to deliver. */ + if (unlikely(!(b->sparse_flow_count + b->bulk_flow_count))) + return NULL; + } + +retry: + /* service this class */ + head = &b->decaying_flows; + if (!first_flow || list_empty(head)) { + head = &b->new_flows; + if (list_empty(head)) { + head = &b->old_flows; + if (unlikely(list_empty(head))) { + head = &b->decaying_flows; + if (unlikely(list_empty(head))) + goto begin; + } + } + } + flow = list_first_entry(head, struct cake_flow, flowchain); + q->cur_flow = flow - b->flows; + first_flow = false; + + /* triple isolation (modified DRR++) */ + srchost = &b->hosts[flow->srchost]; + dsthost = &b->hosts[flow->dsthost]; + host_load = 1; + + if (cake_dsrc(q->flow_mode)) + host_load = max(host_load, srchost->srchost_refcnt); + + if (cake_ddst(q->flow_mode)) + host_load = max(host_load, dsthost->dsthost_refcnt); + + WARN_ON(host_load > CAKE_QUEUES); + + /* flow isolation (DRR++) */ + if (flow->deficit <= 0) { + /* The shifted prandom_u32() is a way to apply dithering to + * avoid accumulating roundoff errors + */ + flow->deficit += (b->flow_quantum * quantum_div[host_load] + + (prandom_u32() >> 16)) >> 16; + list_move_tail(&flow->flowchain, &b->old_flows); + + /* Keep all flows with deficits out of the sparse and decaying + * rotations. No non-empty flow can go into the decaying + * rotation, so they can't get deficits + */ + if (flow->set == CAKE_SET_SPARSE) { + if (flow->head) { + b->sparse_flow_count--; + b->bulk_flow_count++; + flow->set = CAKE_SET_BULK; + } else { + /* we've moved it to the bulk rotation for + * correct deficit accounting but we still want + * to count it as a sparse flow, not a bulk one. + */ + flow->set = CAKE_SET_SPARSE_WAIT; + } + } + goto retry; + } + + /* Retrieve a packet via the AQM */ + while (1) { + skb = cake_dequeue_one(sch); + if (!skb) { + /* this queue was actually empty */ + if (cobalt_queue_empty(&flow->cvars, &b->cparams, now)) + b->unresponsive_flow_count--; + + if (flow->cvars.p_drop || flow->cvars.count || + ktime_before(now, flow->cvars.drop_next)) { + /* keep in the flowchain until the state has + * decayed to rest + */ + list_move_tail(&flow->flowchain, + &b->decaying_flows); + if (flow->set == CAKE_SET_BULK) { + b->bulk_flow_count--; + b->decaying_flow_count++; + } else if (flow->set == CAKE_SET_SPARSE || + flow->set == CAKE_SET_SPARSE_WAIT) { + b->sparse_flow_count--; + b->decaying_flow_count++; + } + flow->set = CAKE_SET_DECAYING; + } else { + /* remove empty queue from the flowchain */ + list_del_init(&flow->flowchain); + if (flow->set == CAKE_SET_SPARSE || + flow->set == CAKE_SET_SPARSE_WAIT) + b->sparse_flow_count--; + else if (flow->set == CAKE_SET_BULK) + b->bulk_flow_count--; + else + b->decaying_flow_count--; + + flow->set = CAKE_SET_NONE; + srchost->srchost_refcnt--; + dsthost->dsthost_refcnt--; + } + goto begin; + } + + /* Last packet in queue may be marked, shouldn't be dropped */ + if (!cobalt_should_drop(&flow->cvars, &b->cparams, now, skb, + (b->bulk_flow_count * + !!(q->rate_flags & + CAKE_FLAG_INGRESS))) || + !flow->head) + break; + + /* drop this packet, get another one */ + if (q->rate_flags & CAKE_FLAG_INGRESS) { + len = cake_advance_shaper(q, b, skb, + now, true); + flow->deficit -= len; + b->tin_deficit -= len; + } + flow->dropped++; + b->tin_dropped++; + qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb)); + qdisc_qstats_drop(sch); + kfree_skb(skb); + if (q->rate_flags & CAKE_FLAG_INGRESS) + goto retry; + } + + b->tin_ecn_mark += !!flow->cvars.ecn_marked; + qdisc_bstats_update(sch, skb); + + /* collect delay stats */ + delay = ktime_to_ns(ktime_sub(now, cobalt_get_enqueue_time(skb))); + b->avge_delay = cake_ewma(b->avge_delay, delay, 8); + b->peak_delay = cake_ewma(b->peak_delay, delay, + delay > b->peak_delay ? 2 : 8); + b->base_delay = cake_ewma(b->base_delay, delay, + delay < b->base_delay ? 2 : 8); + + len = cake_advance_shaper(q, b, skb, now, false); + flow->deficit -= len; + b->tin_deficit -= len; + + if (ktime_after(q->time_next_packet, now) && sch->q.qlen) { + u64 next = min(ktime_to_ns(q->time_next_packet), + ktime_to_ns(q->failsafe_next_packet)); + + qdisc_watchdog_schedule_ns(&q->watchdog, next); + } else if (!sch->q.qlen) { + int i; + + for (i = 0; i < q->tin_cnt; i++) { + if (q->tins[i].decaying_flow_count) { + ktime_t next = \ + ktime_add_ns(now, + q->tins[i].cparams.target); + + qdisc_watchdog_schedule_ns(&q->watchdog, + ktime_to_ns(next)); + break; + } + } + } + + if (q->overflow_timeout) + q->overflow_timeout--; + + return skb; +} + +static void cake_reset(struct Qdisc *sch) +{ + u32 c; + + for (c = 0; c < CAKE_MAX_TINS; c++) + cake_clear_tin(sch, c); +} + +static const struct nla_policy cake_policy[TCA_CAKE_MAX + 1] = { + [TCA_CAKE_BASE_RATE64] = { .type = NLA_U64 }, + [TCA_CAKE_DIFFSERV_MODE] = { .type = NLA_U32 }, + [TCA_CAKE_ATM] = { .type = NLA_U32 }, + [TCA_CAKE_FLOW_MODE] = { .type = NLA_U32 }, + [TCA_CAKE_OVERHEAD] = { .type = NLA_S32 }, + [TCA_CAKE_RTT] = { .type = NLA_U32 }, + [TCA_CAKE_TARGET] = { .type = NLA_U32 }, + [TCA_CAKE_AUTORATE] = { .type = NLA_U32 }, + [TCA_CAKE_MEMORY] = { .type = NLA_U32 }, + [TCA_CAKE_NAT] = { .type = NLA_U32 }, + [TCA_CAKE_RAW] = { .type = NLA_U32 }, + [TCA_CAKE_WASH] = { .type = NLA_U32 }, + [TCA_CAKE_MPU] = { .type = NLA_U32 }, + [TCA_CAKE_INGRESS] = { .type = NLA_U32 }, + [TCA_CAKE_ACK_FILTER] = { .type = NLA_U32 }, +}; + +static void cake_set_rate(struct cake_tin_data *b, u64 rate, u32 mtu, + u64 target_ns, u64 rtt_est_ns) +{ + /* convert byte-rate into time-per-byte + * so it will always unwedge in reasonable time. + */ + static const u64 MIN_RATE = 64; + u32 byte_target = mtu; + u64 byte_target_ns; + u8 rate_shft = 0; + u64 rate_ns = 0; + + b->flow_quantum = 1514; + if (rate) { + b->flow_quantum = max(min(rate >> 12, 1514ULL), 300ULL); + rate_shft = 34; + rate_ns = ((u64)NSEC_PER_SEC) << rate_shft; + rate_ns = div64_u64(rate_ns, max(MIN_RATE, rate)); + while (!!(rate_ns >> 34)) { + rate_ns >>= 1; + rate_shft--; + } + } /* else unlimited, ie. zero delay */ + + b->tin_rate_bps = rate; + b->tin_rate_ns = rate_ns; + b->tin_rate_shft = rate_shft; + + byte_target_ns = (byte_target * rate_ns) >> rate_shft; + + b->cparams.target = max((byte_target_ns * 3) / 2, target_ns); + b->cparams.interval = max(rtt_est_ns + + b->cparams.target - target_ns, + b->cparams.target * 2); + b->cparams.mtu_time = byte_target_ns; + b->cparams.p_inc = 1 << 24; /* 1/256 */ + b->cparams.p_dec = 1 << 20; /* 1/4096 */ +} + +static int cake_config_besteffort(struct Qdisc *sch) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct cake_tin_data *b = &q->tins[0]; + u32 mtu = psched_mtu(qdisc_dev(sch)); + u64 rate = q->rate_bps; + + q->tin_cnt = 1; + + q->tin_index = besteffort; + q->tin_order = normal_order; + + cake_set_rate(b, rate, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + b->tin_quantum_band = 65535; + b->tin_quantum_prio = 65535; + + return 0; +} + +static int cake_config_precedence(struct Qdisc *sch) +{ + /* convert high-level (user visible) parameters into internal format */ + struct cake_sched_data *q = qdisc_priv(sch); + u32 mtu = psched_mtu(qdisc_dev(sch)); + u64 rate = q->rate_bps; + u32 quantum1 = 256; + u32 quantum2 = 256; + u32 i; + + q->tin_cnt = 8; + q->tin_index = precedence; + q->tin_order = normal_order; + + for (i = 0; i < q->tin_cnt; i++) { + struct cake_tin_data *b = &q->tins[i]; + + cake_set_rate(b, rate, mtu, us_to_ns(q->target), + us_to_ns(q->interval)); + + b->tin_quantum_prio = max_t(u16, 1U, quantum1); + b->tin_quantum_band = max_t(u16, 1U, quantum2); + + /* calculate next class's parameters */ + rate *= 7; + rate >>= 3; + + quantum1 *= 3; + quantum1 >>= 1; + + quantum2 *= 7; + quantum2 >>= 3; + } + + return 0; +} + +/* List of known Diffserv codepoints: + * + * Least Effort (CS1) + * Best Effort (CS0) + * Max Reliability & LLT "Lo" (TOS1) + * Max Throughput (TOS2) + * Min Delay (TOS4) + * LLT "La" (TOS5) + * Assured Forwarding 1 (AF1x) - x3 + * Assured Forwarding 2 (AF2x) - x3 + * Assured Forwarding 3 (AF3x) - x3 + * Assured Forwarding 4 (AF4x) - x3 + * Precedence Class 2 (CS2) + * Precedence Class 3 (CS3) + * Precedence Class 4 (CS4) + * Precedence Class 5 (CS5) + * Precedence Class 6 (CS6) + * Precedence Class 7 (CS7) + * Voice Admit (VA) + * Expedited Forwarding (EF) + + * Total 25 codepoints. + */ + +/* List of traffic classes in RFC 4594: + * (roughly descending order of contended priority) + * (roughly ascending order of uncontended throughput) + * + * Network Control (CS6,CS7) - routing traffic + * Telephony (EF,VA) - aka. VoIP streams + * Signalling (CS5) - VoIP setup + * Multimedia Conferencing (AF4x) - aka. video calls + * Realtime Interactive (CS4) - eg. games + * Multimedia Streaming (AF3x) - eg. YouTube, NetFlix, Twitch + * Broadcast Video (CS3) + * Low Latency Data (AF2x,TOS4) - eg. database + * Ops, Admin, Management (CS2,TOS1) - eg. ssh + * Standard Service (CS0 & unrecognised codepoints) + * High Throughput Data (AF1x,TOS2) - eg. web traffic + * Low Priority Data (CS1) - eg. BitTorrent + + * Total 12 traffic classes. + */ + +static int cake_config_diffserv8(struct Qdisc *sch) +{ +/* Pruned list of traffic classes for typical applications: + * + * Network Control (CS6, CS7) + * Minimum Latency (EF, VA, CS5, CS4) + * Interactive Shell (CS2, TOS1) + * Low Latency Transactions (AF2x, TOS4) + * Video Streaming (AF4x, AF3x, CS3) + * Bog Standard (CS0 etc.) + * High Throughput (AF1x, TOS2) + * Background Traffic (CS1) + * + * Total 8 traffic classes. + */ + + struct cake_sched_data *q = qdisc_priv(sch); + u32 mtu = psched_mtu(qdisc_dev(sch)); + u64 rate = q->rate_bps; + u32 quantum1 = 256; + u32 quantum2 = 256; + u32 i; + + q->tin_cnt = 8; + + /* codepoint to class mapping */ + q->tin_index = diffserv8; + q->tin_order = normal_order; + + /* class characteristics */ + for (i = 0; i < q->tin_cnt; i++) { + struct cake_tin_data *b = &q->tins[i]; + + cake_set_rate(b, rate, mtu, us_to_ns(q->target), + us_to_ns(q->interval)); + + b->tin_quantum_prio = max_t(u16, 1U, quantum1); + b->tin_quantum_band = max_t(u16, 1U, quantum2); + + /* calculate next class's parameters */ + rate *= 7; + rate >>= 3; + + quantum1 *= 3; + quantum1 >>= 1; + + quantum2 *= 7; + quantum2 >>= 3; + } + + return 0; +} + +static int cake_config_diffserv4(struct Qdisc *sch) +{ +/* Further pruned list of traffic classes for four-class system: + * + * Latency Sensitive (CS7, CS6, EF, VA, CS5, CS4) + * Streaming Media (AF4x, AF3x, CS3, AF2x, TOS4, CS2, TOS1) + * Best Effort (CS0, AF1x, TOS2, and those not specified) + * Background Traffic (CS1) + * + * Total 4 traffic classes. + */ + + struct cake_sched_data *q = qdisc_priv(sch); + u32 mtu = psched_mtu(qdisc_dev(sch)); + u64 rate = q->rate_bps; + u32 quantum = 1024; + + q->tin_cnt = 4; + + /* codepoint to class mapping */ + q->tin_index = diffserv4; + q->tin_order = bulk_order; + + /* class characteristics */ + cake_set_rate(&q->tins[0], rate, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + cake_set_rate(&q->tins[1], rate >> 4, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + cake_set_rate(&q->tins[2], rate >> 1, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + cake_set_rate(&q->tins[3], rate >> 2, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + + /* priority weights */ + q->tins[0].tin_quantum_prio = quantum; + q->tins[1].tin_quantum_prio = quantum >> 4; + q->tins[2].tin_quantum_prio = quantum << 2; + q->tins[3].tin_quantum_prio = quantum << 4; + + /* bandwidth-sharing weights */ + q->tins[0].tin_quantum_band = quantum; + q->tins[1].tin_quantum_band = quantum >> 4; + q->tins[2].tin_quantum_band = quantum >> 1; + q->tins[3].tin_quantum_band = quantum >> 2; + + return 0; +} + +static int cake_config_diffserv3(struct Qdisc *sch) +{ +/* Simplified Diffserv structure with 3 tins. + * Low Priority (CS1) + * Best Effort + * Latency Sensitive (TOS4, VA, EF, CS6, CS7) + */ + struct cake_sched_data *q = qdisc_priv(sch); + u32 mtu = psched_mtu(qdisc_dev(sch)); + u64 rate = q->rate_bps; + u32 quantum = 1024; + + q->tin_cnt = 3; + + /* codepoint to class mapping */ + q->tin_index = diffserv3; + q->tin_order = bulk_order; + + /* class characteristics */ + cake_set_rate(&q->tins[0], rate, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + cake_set_rate(&q->tins[1], rate >> 4, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + cake_set_rate(&q->tins[2], rate >> 2, mtu, + us_to_ns(q->target), us_to_ns(q->interval)); + + /* priority weights */ + q->tins[0].tin_quantum_prio = quantum; + q->tins[1].tin_quantum_prio = quantum >> 4; + q->tins[2].tin_quantum_prio = quantum << 4; + + /* bandwidth-sharing weights */ + q->tins[0].tin_quantum_band = quantum; + q->tins[1].tin_quantum_band = quantum >> 4; + q->tins[2].tin_quantum_band = quantum >> 2; + + return 0; +} + +static void cake_reconfigure(struct Qdisc *sch) +{ + struct cake_sched_data *q = qdisc_priv(sch); + int c, ft; + + switch (q->tin_mode) { + case CAKE_DIFFSERV_BESTEFFORT: + ft = cake_config_besteffort(sch); + break; + + case CAKE_DIFFSERV_PRECEDENCE: + ft = cake_config_precedence(sch); + break; + + case CAKE_DIFFSERV_DIFFSERV8: + ft = cake_config_diffserv8(sch); + break; + + case CAKE_DIFFSERV_DIFFSERV4: + ft = cake_config_diffserv4(sch); + break; + + case CAKE_DIFFSERV_DIFFSERV3: + default: + ft = cake_config_diffserv3(sch); + break; + } + + for (c = q->tin_cnt; c < CAKE_MAX_TINS; c++) { + cake_clear_tin(sch, c); + q->tins[c].cparams.mtu_time = q->tins[ft].cparams.mtu_time; + } + + q->rate_ns = q->tins[ft].tin_rate_ns; + q->rate_shft = q->tins[ft].tin_rate_shft; + + if (q->buffer_config_limit) { + q->buffer_limit = q->buffer_config_limit; + } else if (q->rate_bps) { + u64 t = q->rate_bps * q->interval; + + do_div(t, USEC_PER_SEC / 4); + q->buffer_limit = max_t(u32, t, 4U << 20); + } else { + q->buffer_limit = ~0; + } + + sch->flags &= ~TCQ_F_CAN_BYPASS; + + q->buffer_limit = min(q->buffer_limit, + max(sch->limit * psched_mtu(qdisc_dev(sch)), + q->buffer_config_limit)); +} + +static int cake_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_CAKE_MAX + 1]; + int err; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_CAKE_MAX, opt, cake_policy, extack); + if (err < 0) + return err; + + if (tb[TCA_CAKE_NAT]) { +#if IS_ENABLED(CONFIG_NF_CONNTRACK) + q->flow_mode &= ~CAKE_FLOW_NAT_FLAG; + q->flow_mode |= CAKE_FLOW_NAT_FLAG * + !!nla_get_u32(tb[TCA_CAKE_NAT]); +#else + NL_SET_ERR_MSG_ATTR(extack, tb[TCA_CAKE_NAT], + "No conntrack support in kernel"); + return -EOPNOTSUPP; +#endif + } + + if (tb[TCA_CAKE_BASE_RATE64]) + q->rate_bps = nla_get_u64(tb[TCA_CAKE_BASE_RATE64]); + + if (tb[TCA_CAKE_DIFFSERV_MODE]) + q->tin_mode = nla_get_u32(tb[TCA_CAKE_DIFFSERV_MODE]); + + if (tb[TCA_CAKE_WASH]) { + if (!!nla_get_u32(tb[TCA_CAKE_WASH])) + q->rate_flags |= CAKE_FLAG_WASH; + else + q->rate_flags &= ~CAKE_FLAG_WASH; + } + + if (tb[TCA_CAKE_FLOW_MODE]) + q->flow_mode = ((q->flow_mode & CAKE_FLOW_NAT_FLAG) | + (nla_get_u32(tb[TCA_CAKE_FLOW_MODE]) & + CAKE_FLOW_MASK)); + + if (tb[TCA_CAKE_ATM]) + q->atm_mode = nla_get_u32(tb[TCA_CAKE_ATM]); + + if (tb[TCA_CAKE_OVERHEAD]) { + q->rate_overhead = nla_get_s32(tb[TCA_CAKE_OVERHEAD]); + q->rate_flags |= CAKE_FLAG_OVERHEAD; + + q->max_netlen = 0; + q->max_adjlen = 0; + q->min_netlen = ~0; + q->min_adjlen = ~0; + } + + if (tb[TCA_CAKE_RAW]) { + q->rate_flags &= ~CAKE_FLAG_OVERHEAD; + + q->max_netlen = 0; + q->max_adjlen = 0; + q->min_netlen = ~0; + q->min_adjlen = ~0; + } + + if (tb[TCA_CAKE_MPU]) + q->rate_mpu = nla_get_u32(tb[TCA_CAKE_MPU]); + + if (tb[TCA_CAKE_RTT]) { + q->interval = nla_get_u32(tb[TCA_CAKE_RTT]); + + if (!q->interval) + q->interval = 1; + } + + if (tb[TCA_CAKE_TARGET]) { + q->target = nla_get_u32(tb[TCA_CAKE_TARGET]); + + if (!q->target) + q->target = 1; + } + + if (tb[TCA_CAKE_AUTORATE]) { + if (!!nla_get_u32(tb[TCA_CAKE_AUTORATE])) + q->rate_flags |= CAKE_FLAG_AUTORATE_INGRESS; + else + q->rate_flags &= ~CAKE_FLAG_AUTORATE_INGRESS; + } + + if (tb[TCA_CAKE_INGRESS]) { + if (!!nla_get_u32(tb[TCA_CAKE_INGRESS])) + q->rate_flags |= CAKE_FLAG_INGRESS; + else + q->rate_flags &= ~CAKE_FLAG_INGRESS; + } + + if (tb[TCA_CAKE_ACK_FILTER]) + q->ack_filter = nla_get_u32(tb[TCA_CAKE_ACK_FILTER]); + + if (tb[TCA_CAKE_MEMORY]) + q->buffer_config_limit = nla_get_u32(tb[TCA_CAKE_MEMORY]); + + if (tb[TCA_CAKE_SPLIT_GSO]) { + if (!!nla_get_u32(tb[TCA_CAKE_SPLIT_GSO])) + q->rate_flags |= CAKE_FLAG_SPLIT_GSO; + else + q->rate_flags &= ~CAKE_FLAG_SPLIT_GSO; + } + + if (q->tins) { + sch_tree_lock(sch); + cake_reconfigure(sch); + sch_tree_unlock(sch); + } + + return 0; +} + +static void cake_destroy(struct Qdisc *sch) +{ + struct cake_sched_data *q = qdisc_priv(sch); + + qdisc_watchdog_cancel(&q->watchdog); + tcf_block_put(q->block); + kvfree(q->tins); +} + +static int cake_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct cake_sched_data *q = qdisc_priv(sch); + int i, j, err; + + sch->limit = 10240; + q->tin_mode = CAKE_DIFFSERV_DIFFSERV3; + q->flow_mode = CAKE_FLOW_TRIPLE; + + q->rate_bps = 0; /* unlimited by default */ + + q->interval = 100000; /* 100ms default */ + q->target = 5000; /* 5ms: codel RFC argues + * for 5 to 10% of interval + */ + q->rate_flags |= CAKE_FLAG_SPLIT_GSO; + q->cur_tin = 0; + q->cur_flow = 0; + + qdisc_watchdog_init(&q->watchdog, sch); + + if (opt) { + err = cake_change(sch, opt, extack); + + if (err) + return err; + } + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + quantum_div[0] = ~0; + for (i = 1; i <= CAKE_QUEUES; i++) + quantum_div[i] = 65535 / i; + + q->tins = kvcalloc(CAKE_MAX_TINS, sizeof(struct cake_tin_data), + GFP_KERNEL); + if (!q->tins) + return -ENOMEM; + + for (i = 0; i < CAKE_MAX_TINS; i++) { + struct cake_tin_data *b = q->tins + i; + + INIT_LIST_HEAD(&b->new_flows); + INIT_LIST_HEAD(&b->old_flows); + INIT_LIST_HEAD(&b->decaying_flows); + b->sparse_flow_count = 0; + b->bulk_flow_count = 0; + b->decaying_flow_count = 0; + + for (j = 0; j < CAKE_QUEUES; j++) { + struct cake_flow *flow = b->flows + j; + u32 k = j * CAKE_MAX_TINS + i; + + INIT_LIST_HEAD(&flow->flowchain); + cobalt_vars_init(&flow->cvars); + + q->overflow_heap[k].t = i; + q->overflow_heap[k].b = j; + b->overflow_idx[j] = k; + } + } + + cake_reconfigure(sch); + q->avg_peak_bandwidth = q->rate_bps; + q->min_netlen = ~0; + q->min_adjlen = ~0; + return 0; +} + +static int cake_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct cake_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (!opts) + goto nla_put_failure; + + if (nla_put_u64_64bit(skb, TCA_CAKE_BASE_RATE64, q->rate_bps, + TCA_CAKE_PAD)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_FLOW_MODE, + q->flow_mode & CAKE_FLOW_MASK)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_RTT, q->interval)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_TARGET, q->target)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_MEMORY, q->buffer_config_limit)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_AUTORATE, + !!(q->rate_flags & CAKE_FLAG_AUTORATE_INGRESS))) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_INGRESS, + !!(q->rate_flags & CAKE_FLAG_INGRESS))) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_ACK_FILTER, q->ack_filter)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_NAT, + !!(q->flow_mode & CAKE_FLOW_NAT_FLAG))) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_DIFFSERV_MODE, q->tin_mode)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_WASH, + !!(q->rate_flags & CAKE_FLAG_WASH))) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_OVERHEAD, q->rate_overhead)) + goto nla_put_failure; + + if (!(q->rate_flags & CAKE_FLAG_OVERHEAD)) + if (nla_put_u32(skb, TCA_CAKE_RAW, 0)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_ATM, q->atm_mode)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_MPU, q->rate_mpu)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CAKE_SPLIT_GSO, + !!(q->rate_flags & CAKE_FLAG_SPLIT_GSO))) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + return -1; +} + +static int cake_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct nlattr *stats = nla_nest_start(d->skb, TCA_STATS_APP); + struct cake_sched_data *q = qdisc_priv(sch); + struct nlattr *tstats, *ts; + int i; + + if (!stats) + return -1; + +#define PUT_STAT_U32(attr, data) do { \ + if (nla_put_u32(d->skb, TCA_CAKE_STATS_ ## attr, data)) \ + goto nla_put_failure; \ + } while (0) +#define PUT_STAT_U64(attr, data) do { \ + if (nla_put_u64_64bit(d->skb, TCA_CAKE_STATS_ ## attr, \ + data, TCA_CAKE_STATS_PAD)) \ + goto nla_put_failure; \ + } while (0) + + PUT_STAT_U64(CAPACITY_ESTIMATE64, q->avg_peak_bandwidth); + PUT_STAT_U32(MEMORY_LIMIT, q->buffer_limit); + PUT_STAT_U32(MEMORY_USED, q->buffer_max_used); + PUT_STAT_U32(AVG_NETOFF, ((q->avg_netoff + 0x8000) >> 16)); + PUT_STAT_U32(MAX_NETLEN, q->max_netlen); + PUT_STAT_U32(MAX_ADJLEN, q->max_adjlen); + PUT_STAT_U32(MIN_NETLEN, q->min_netlen); + PUT_STAT_U32(MIN_ADJLEN, q->min_adjlen); + +#undef PUT_STAT_U32 +#undef PUT_STAT_U64 + + tstats = nla_nest_start(d->skb, TCA_CAKE_STATS_TIN_STATS); + if (!tstats) + goto nla_put_failure; + +#define PUT_TSTAT_U32(attr, data) do { \ + if (nla_put_u32(d->skb, TCA_CAKE_TIN_STATS_ ## attr, data)) \ + goto nla_put_failure; \ + } while (0) +#define PUT_TSTAT_U64(attr, data) do { \ + if (nla_put_u64_64bit(d->skb, TCA_CAKE_TIN_STATS_ ## attr, \ + data, TCA_CAKE_TIN_STATS_PAD)) \ + goto nla_put_failure; \ + } while (0) + + for (i = 0; i < q->tin_cnt; i++) { + struct cake_tin_data *b = &q->tins[q->tin_order[i]]; + + ts = nla_nest_start(d->skb, i + 1); + if (!ts) + goto nla_put_failure; + + PUT_TSTAT_U64(THRESHOLD_RATE64, b->tin_rate_bps); + PUT_TSTAT_U64(SENT_BYTES64, b->bytes); + PUT_TSTAT_U32(BACKLOG_BYTES, b->tin_backlog); + + PUT_TSTAT_U32(TARGET_US, + ktime_to_us(ns_to_ktime(b->cparams.target))); + PUT_TSTAT_U32(INTERVAL_US, + ktime_to_us(ns_to_ktime(b->cparams.interval))); + + PUT_TSTAT_U32(SENT_PACKETS, b->packets); + PUT_TSTAT_U32(DROPPED_PACKETS, b->tin_dropped); + PUT_TSTAT_U32(ECN_MARKED_PACKETS, b->tin_ecn_mark); + PUT_TSTAT_U32(ACKS_DROPPED_PACKETS, b->ack_drops); + + PUT_TSTAT_U32(PEAK_DELAY_US, + ktime_to_us(ns_to_ktime(b->peak_delay))); + PUT_TSTAT_U32(AVG_DELAY_US, + ktime_to_us(ns_to_ktime(b->avge_delay))); + PUT_TSTAT_U32(BASE_DELAY_US, + ktime_to_us(ns_to_ktime(b->base_delay))); + + PUT_TSTAT_U32(WAY_INDIRECT_HITS, b->way_hits); + PUT_TSTAT_U32(WAY_MISSES, b->way_misses); + PUT_TSTAT_U32(WAY_COLLISIONS, b->way_collisions); + + PUT_TSTAT_U32(SPARSE_FLOWS, b->sparse_flow_count + + b->decaying_flow_count); + PUT_TSTAT_U32(BULK_FLOWS, b->bulk_flow_count); + PUT_TSTAT_U32(UNRESPONSIVE_FLOWS, b->unresponsive_flow_count); + PUT_TSTAT_U32(MAX_SKBLEN, b->max_skblen); + + PUT_TSTAT_U32(FLOW_QUANTUM, b->flow_quantum); + nla_nest_end(d->skb, ts); + } + +#undef PUT_TSTAT_U32 +#undef PUT_TSTAT_U64 + + nla_nest_end(d->skb, tstats); + return nla_nest_end(d->skb, stats); + +nla_put_failure: + nla_nest_cancel(d->skb, stats); + return -1; +} + +static struct Qdisc *cake_leaf(struct Qdisc *sch, unsigned long arg) +{ + return NULL; +} + +static unsigned long cake_find(struct Qdisc *sch, u32 classid) +{ + return 0; +} + +static unsigned long cake_bind(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + return 0; +} + +static void cake_unbind(struct Qdisc *q, unsigned long cl) +{ +} + +static struct tcf_block *cake_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct cake_sched_data *q = qdisc_priv(sch); + + if (cl) + return NULL; + return q->block; +} + +static int cake_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + tcm->tcm_handle |= TC_H_MIN(cl); + return 0; +} + +static int cake_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct cake_sched_data *q = qdisc_priv(sch); + const struct cake_flow *flow = NULL; + struct gnet_stats_queue qs = { 0 }; + struct nlattr *stats; + u32 idx = cl - 1; + + if (idx < CAKE_QUEUES * q->tin_cnt) { + const struct cake_tin_data *b = \ + &q->tins[q->tin_order[idx / CAKE_QUEUES]]; + const struct sk_buff *skb; + + flow = &b->flows[idx % CAKE_QUEUES]; + + if (flow->head) { + sch_tree_lock(sch); + skb = flow->head; + while (skb) { + qs.qlen++; + skb = skb->next; + } + sch_tree_unlock(sch); + } + qs.backlog = b->backlogs[idx % CAKE_QUEUES]; + qs.drops = flow->dropped; + } + if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0) + return -1; + if (flow) { + ktime_t now = ktime_get(); + + stats = nla_nest_start(d->skb, TCA_STATS_APP); + if (!stats) + return -1; + +#define PUT_STAT_U32(attr, data) do { \ + if (nla_put_u32(d->skb, TCA_CAKE_STATS_ ## attr, data)) \ + goto nla_put_failure; \ + } while (0) +#define PUT_STAT_S32(attr, data) do { \ + if (nla_put_s32(d->skb, TCA_CAKE_STATS_ ## attr, data)) \ + goto nla_put_failure; \ + } while (0) + + PUT_STAT_S32(DEFICIT, flow->deficit); + PUT_STAT_U32(DROPPING, flow->cvars.dropping); + PUT_STAT_U32(COBALT_COUNT, flow->cvars.count); + PUT_STAT_U32(P_DROP, flow->cvars.p_drop); + if (flow->cvars.p_drop) { + PUT_STAT_S32(BLUE_TIMER_US, + ktime_to_us( + ktime_sub(now, + flow->cvars.blue_timer))); + } + if (flow->cvars.dropping) { + PUT_STAT_S32(DROP_NEXT_US, + ktime_to_us( + ktime_sub(now, + flow->cvars.drop_next))); + } + + if (nla_nest_end(d->skb, stats) < 0) + return -1; + } + + return 0; + +nla_put_failure: + nla_nest_cancel(d->skb, stats); + return -1; +} + +static void cake_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct cake_sched_data *q = qdisc_priv(sch); + unsigned int i, j; + + if (arg->stop) + return; + + for (i = 0; i < q->tin_cnt; i++) { + struct cake_tin_data *b = &q->tins[q->tin_order[i]]; + + for (j = 0; j < CAKE_QUEUES; j++) { + if (list_empty(&b->flows[j].flowchain) || + arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, i * CAKE_QUEUES + j + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } + } +} + +static const struct Qdisc_class_ops cake_class_ops = { + .leaf = cake_leaf, + .find = cake_find, + .tcf_block = cake_tcf_block, + .bind_tcf = cake_bind, + .unbind_tcf = cake_unbind, + .dump = cake_dump_class, + .dump_stats = cake_dump_class_stats, + .walk = cake_walk, +}; + +static struct Qdisc_ops cake_qdisc_ops __read_mostly = { + .cl_ops = &cake_class_ops, + .id = "cake", + .priv_size = sizeof(struct cake_sched_data), + .enqueue = cake_enqueue, + .dequeue = cake_dequeue, + .peek = qdisc_peek_dequeued, + .init = cake_init, + .reset = cake_reset, + .destroy = cake_destroy, + .change = cake_change, + .dump = cake_dump, + .dump_stats = cake_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init cake_module_init(void) +{ + return register_qdisc(&cake_qdisc_ops); +} + +static void __exit cake_module_exit(void) +{ + unregister_qdisc(&cake_qdisc_ops); +} + +module_init(cake_module_init) +module_exit(cake_module_exit) +MODULE_AUTHOR("Jonathan Morton"); +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_DESCRIPTION("The CAKE shaper."); diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c new file mode 100644 index 000000000..0a76ad05e --- /dev/null +++ b/net/sched/sch_cbq.c @@ -0,0 +1,1823 @@ +/* + * net/sched/sch_cbq.c Class-Based Queueing discipline. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + + +/* Class-Based Queueing (CBQ) algorithm. + ======================================= + + Sources: [1] Sally Floyd and Van Jacobson, "Link-sharing and Resource + Management Models for Packet Networks", + IEEE/ACM Transactions on Networking, Vol.3, No.4, 1995 + + [2] Sally Floyd, "Notes on CBQ and Guaranteed Service", 1995 + + [3] Sally Floyd, "Notes on Class-Based Queueing: Setting + Parameters", 1996 + + [4] Sally Floyd and Michael Speer, "Experimental Results + for Class-Based Queueing", 1998, not published. + + ----------------------------------------------------------------------- + + Algorithm skeleton was taken from NS simulator cbq.cc. + If someone wants to check this code against the LBL version, + he should take into account that ONLY the skeleton was borrowed, + the implementation is different. Particularly: + + --- The WRR algorithm is different. Our version looks more + reasonable (I hope) and works when quanta are allowed to be + less than MTU, which is always the case when real time classes + have small rates. Note, that the statement of [3] is + incomplete, delay may actually be estimated even if class + per-round allotment is less than MTU. Namely, if per-round + allotment is W*r_i, and r_1+...+r_k = r < 1 + + delay_i <= ([MTU/(W*r_i)]*W*r + W*r + k*MTU)/B + + In the worst case we have IntServ estimate with D = W*r+k*MTU + and C = MTU*r. The proof (if correct at all) is trivial. + + + --- It seems that cbq-2.0 is not very accurate. At least, I cannot + interpret some places, which look like wrong translations + from NS. Anyone is advised to find these differences + and explain to me, why I am wrong 8). + + --- Linux has no EOI event, so that we cannot estimate true class + idle time. Workaround is to consider the next dequeue event + as sign that previous packet is finished. This is wrong because of + internal device queueing, but on a permanently loaded link it is true. + Moreover, combined with clock integrator, this scheme looks + very close to an ideal solution. */ + +struct cbq_sched_data; + + +struct cbq_class { + struct Qdisc_class_common common; + struct cbq_class *next_alive; /* next class with backlog in this priority band */ + +/* Parameters */ + unsigned char priority; /* class priority */ + unsigned char priority2; /* priority to be used after overlimit */ + unsigned char ewma_log; /* time constant for idle time calculation */ + + u32 defmap; + + /* Link-sharing scheduler parameters */ + long maxidle; /* Class parameters: see below. */ + long offtime; + long minidle; + u32 avpkt; + struct qdisc_rate_table *R_tab; + + /* General scheduler (WRR) parameters */ + long allot; + long quantum; /* Allotment per WRR round */ + long weight; /* Relative allotment: see below */ + + struct Qdisc *qdisc; /* Ptr to CBQ discipline */ + struct cbq_class *split; /* Ptr to split node */ + struct cbq_class *share; /* Ptr to LS parent in the class tree */ + struct cbq_class *tparent; /* Ptr to tree parent in the class tree */ + struct cbq_class *borrow; /* NULL if class is bandwidth limited; + parent otherwise */ + struct cbq_class *sibling; /* Sibling chain */ + struct cbq_class *children; /* Pointer to children chain */ + + struct Qdisc *q; /* Elementary queueing discipline */ + + +/* Variables */ + unsigned char cpriority; /* Effective priority */ + unsigned char delayed; + unsigned char level; /* level of the class in hierarchy: + 0 for leaf classes, and maximal + level of children + 1 for nodes. + */ + + psched_time_t last; /* Last end of service */ + psched_time_t undertime; + long avgidle; + long deficit; /* Saved deficit for WRR */ + psched_time_t penalized; + struct gnet_stats_basic_packed bstats; + struct gnet_stats_queue qstats; + struct net_rate_estimator __rcu *rate_est; + struct tc_cbq_xstats xstats; + + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + + int filters; + + struct cbq_class *defaults[TC_PRIO_MAX + 1]; +}; + +struct cbq_sched_data { + struct Qdisc_class_hash clhash; /* Hash table of all classes */ + int nclasses[TC_CBQ_MAXPRIO + 1]; + unsigned int quanta[TC_CBQ_MAXPRIO + 1]; + + struct cbq_class link; + + unsigned int activemask; + struct cbq_class *active[TC_CBQ_MAXPRIO + 1]; /* List of all classes + with backlog */ + +#ifdef CONFIG_NET_CLS_ACT + struct cbq_class *rx_class; +#endif + struct cbq_class *tx_class; + struct cbq_class *tx_borrowed; + int tx_len; + psched_time_t now; /* Cached timestamp */ + unsigned int pmask; + + struct hrtimer delay_timer; + struct qdisc_watchdog watchdog; /* Watchdog timer, + started when CBQ has + backlog, but cannot + transmit just now */ + psched_tdiff_t wd_expires; + int toplevel; + u32 hgenerator; +}; + + +#define L2T(cl, len) qdisc_l2t((cl)->R_tab, len) + +static inline struct cbq_class * +cbq_class_lookup(struct cbq_sched_data *q, u32 classid) +{ + struct Qdisc_class_common *clc; + + clc = qdisc_class_find(&q->clhash, classid); + if (clc == NULL) + return NULL; + return container_of(clc, struct cbq_class, common); +} + +#ifdef CONFIG_NET_CLS_ACT + +static struct cbq_class * +cbq_reclassify(struct sk_buff *skb, struct cbq_class *this) +{ + struct cbq_class *cl; + + for (cl = this->tparent; cl; cl = cl->tparent) { + struct cbq_class *new = cl->defaults[TC_PRIO_BESTEFFORT]; + + if (new != NULL && new != this) + return new; + } + return NULL; +} + +#endif + +/* Classify packet. The procedure is pretty complicated, but + * it allows us to combine link sharing and priority scheduling + * transparently. + * + * Namely, you can put link sharing rules (f.e. route based) at root of CBQ, + * so that it resolves to split nodes. Then packets are classified + * by logical priority, or a more specific classifier may be attached + * to the split node. + */ + +static struct cbq_class * +cbq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *head = &q->link; + struct cbq_class **defmap; + struct cbq_class *cl = NULL; + u32 prio = skb->priority; + struct tcf_proto *fl; + struct tcf_result res; + + /* + * Step 1. If skb->priority points to one of our classes, use it. + */ + if (TC_H_MAJ(prio ^ sch->handle) == 0 && + (cl = cbq_class_lookup(q, prio)) != NULL) + return cl; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + for (;;) { + int result = 0; + defmap = head->defaults; + + fl = rcu_dereference_bh(head->filter_list); + /* + * Step 2+n. Apply classifier. + */ + result = tcf_classify(skb, fl, &res, true); + if (!fl || result < 0) + goto fallback; + + cl = (void *)res.class; + if (!cl) { + if (TC_H_MAJ(res.classid)) + cl = cbq_class_lookup(q, res.classid); + else if ((cl = defmap[res.classid & TC_PRIO_MAX]) == NULL) + cl = defmap[TC_PRIO_BESTEFFORT]; + + if (cl == NULL) + goto fallback; + } + if (cl->level >= head->level) + goto fallback; +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + case TC_ACT_RECLASSIFY: + return cbq_reclassify(skb, cl); + } +#endif + if (cl->level == 0) + return cl; + + /* + * Step 3+n. If classifier selected a link sharing class, + * apply agency specific classifier. + * Repeat this procdure until we hit a leaf node. + */ + head = cl; + } + +fallback: + cl = head; + + /* + * Step 4. No success... + */ + if (TC_H_MAJ(prio) == 0 && + !(cl = head->defaults[prio & TC_PRIO_MAX]) && + !(cl = head->defaults[TC_PRIO_BESTEFFORT])) + return head; + + return cl; +} + +/* + * A packet has just been enqueued on the empty class. + * cbq_activate_class adds it to the tail of active class list + * of its priority band. + */ + +static inline void cbq_activate_class(struct cbq_class *cl) +{ + struct cbq_sched_data *q = qdisc_priv(cl->qdisc); + int prio = cl->cpriority; + struct cbq_class *cl_tail; + + cl_tail = q->active[prio]; + q->active[prio] = cl; + + if (cl_tail != NULL) { + cl->next_alive = cl_tail->next_alive; + cl_tail->next_alive = cl; + } else { + cl->next_alive = cl; + q->activemask |= (1<<prio); + } +} + +/* + * Unlink class from active chain. + * Note that this same procedure is done directly in cbq_dequeue* + * during round-robin procedure. + */ + +static void cbq_deactivate_class(struct cbq_class *this) +{ + struct cbq_sched_data *q = qdisc_priv(this->qdisc); + int prio = this->cpriority; + struct cbq_class *cl; + struct cbq_class *cl_prev = q->active[prio]; + + do { + cl = cl_prev->next_alive; + if (cl == this) { + cl_prev->next_alive = cl->next_alive; + cl->next_alive = NULL; + + if (cl == q->active[prio]) { + q->active[prio] = cl_prev; + if (cl == q->active[prio]) { + q->active[prio] = NULL; + q->activemask &= ~(1<<prio); + return; + } + } + return; + } + } while ((cl_prev = cl) != q->active[prio]); +} + +static void +cbq_mark_toplevel(struct cbq_sched_data *q, struct cbq_class *cl) +{ + int toplevel = q->toplevel; + + if (toplevel > cl->level) { + psched_time_t now = psched_get_time(); + + do { + if (cl->undertime < now) { + q->toplevel = cl->level; + return; + } + } while ((cl = cl->borrow) != NULL && toplevel > cl->level); + } +} + +static int +cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + int uninitialized_var(ret); + struct cbq_class *cl = cbq_classify(skb, sch, &ret); + +#ifdef CONFIG_NET_CLS_ACT + q->rx_class = cl; +#endif + if (cl == NULL) { + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; + } + + ret = qdisc_enqueue(skb, cl->q, to_free); + if (ret == NET_XMIT_SUCCESS) { + sch->q.qlen++; + cbq_mark_toplevel(q, cl); + if (!cl->next_alive) + cbq_activate_class(cl); + return ret; + } + + if (net_xmit_drop_count(ret)) { + qdisc_qstats_drop(sch); + cbq_mark_toplevel(q, cl); + cl->qstats.drops++; + } + return ret; +} + +/* Overlimit action: penalize leaf class by adding offtime */ +static void cbq_overlimit(struct cbq_class *cl) +{ + struct cbq_sched_data *q = qdisc_priv(cl->qdisc); + psched_tdiff_t delay = cl->undertime - q->now; + + if (!cl->delayed) { + delay += cl->offtime; + + /* + * Class goes to sleep, so that it will have no + * chance to work avgidle. Let's forgive it 8) + * + * BTW cbq-2.0 has a crap in this + * place, apparently they forgot to shift it by cl->ewma_log. + */ + if (cl->avgidle < 0) + delay -= (-cl->avgidle) - ((-cl->avgidle) >> cl->ewma_log); + if (cl->avgidle < cl->minidle) + cl->avgidle = cl->minidle; + if (delay <= 0) + delay = 1; + cl->undertime = q->now + delay; + + cl->xstats.overactions++; + cl->delayed = 1; + } + if (q->wd_expires == 0 || q->wd_expires > delay) + q->wd_expires = delay; + + /* Dirty work! We must schedule wakeups based on + * real available rate, rather than leaf rate, + * which may be tiny (even zero). + */ + if (q->toplevel == TC_CBQ_MAXLEVEL) { + struct cbq_class *b; + psched_tdiff_t base_delay = q->wd_expires; + + for (b = cl->borrow; b; b = b->borrow) { + delay = b->undertime - q->now; + if (delay < base_delay) { + if (delay <= 0) + delay = 1; + base_delay = delay; + } + } + + q->wd_expires = base_delay; + } +} + +static psched_tdiff_t cbq_undelay_prio(struct cbq_sched_data *q, int prio, + psched_time_t now) +{ + struct cbq_class *cl; + struct cbq_class *cl_prev = q->active[prio]; + psched_time_t sched = now; + + if (cl_prev == NULL) + return 0; + + do { + cl = cl_prev->next_alive; + if (now - cl->penalized > 0) { + cl_prev->next_alive = cl->next_alive; + cl->next_alive = NULL; + cl->cpriority = cl->priority; + cl->delayed = 0; + cbq_activate_class(cl); + + if (cl == q->active[prio]) { + q->active[prio] = cl_prev; + if (cl == q->active[prio]) { + q->active[prio] = NULL; + return 0; + } + } + + cl = cl_prev->next_alive; + } else if (sched - cl->penalized > 0) + sched = cl->penalized; + } while ((cl_prev = cl) != q->active[prio]); + + return sched - now; +} + +static enum hrtimer_restart cbq_undelay(struct hrtimer *timer) +{ + struct cbq_sched_data *q = container_of(timer, struct cbq_sched_data, + delay_timer); + struct Qdisc *sch = q->watchdog.qdisc; + psched_time_t now; + psched_tdiff_t delay = 0; + unsigned int pmask; + + now = psched_get_time(); + + pmask = q->pmask; + q->pmask = 0; + + while (pmask) { + int prio = ffz(~pmask); + psched_tdiff_t tmp; + + pmask &= ~(1<<prio); + + tmp = cbq_undelay_prio(q, prio, now); + if (tmp > 0) { + q->pmask |= 1<<prio; + if (tmp < delay || delay == 0) + delay = tmp; + } + } + + if (delay) { + ktime_t time; + + time = 0; + time = ktime_add_ns(time, PSCHED_TICKS2NS(now + delay)); + hrtimer_start(&q->delay_timer, time, HRTIMER_MODE_ABS_PINNED); + } + + __netif_schedule(qdisc_root(sch)); + return HRTIMER_NORESTART; +} + +/* + * It is mission critical procedure. + * + * We "regenerate" toplevel cutoff, if transmitting class + * has backlog and it is not regulated. It is not part of + * original CBQ description, but looks more reasonable. + * Probably, it is wrong. This question needs further investigation. + */ + +static inline void +cbq_update_toplevel(struct cbq_sched_data *q, struct cbq_class *cl, + struct cbq_class *borrowed) +{ + if (cl && q->toplevel >= borrowed->level) { + if (cl->q->q.qlen > 1) { + do { + if (borrowed->undertime == PSCHED_PASTPERFECT) { + q->toplevel = borrowed->level; + return; + } + } while ((borrowed = borrowed->borrow) != NULL); + } +#if 0 + /* It is not necessary now. Uncommenting it + will save CPU cycles, but decrease fairness. + */ + q->toplevel = TC_CBQ_MAXLEVEL; +#endif + } +} + +static void +cbq_update(struct cbq_sched_data *q) +{ + struct cbq_class *this = q->tx_class; + struct cbq_class *cl = this; + int len = q->tx_len; + psched_time_t now; + + q->tx_class = NULL; + /* Time integrator. We calculate EOS time + * by adding expected packet transmission time. + */ + now = q->now + L2T(&q->link, len); + + for ( ; cl; cl = cl->share) { + long avgidle = cl->avgidle; + long idle; + + cl->bstats.packets++; + cl->bstats.bytes += len; + + /* + * (now - last) is total time between packet right edges. + * (last_pktlen/rate) is "virtual" busy time, so that + * + * idle = (now - last) - last_pktlen/rate + */ + + idle = now - cl->last; + if ((unsigned long)idle > 128*1024*1024) { + avgidle = cl->maxidle; + } else { + idle -= L2T(cl, len); + + /* true_avgidle := (1-W)*true_avgidle + W*idle, + * where W=2^{-ewma_log}. But cl->avgidle is scaled: + * cl->avgidle == true_avgidle/W, + * hence: + */ + avgidle += idle - (avgidle>>cl->ewma_log); + } + + if (avgidle <= 0) { + /* Overlimit or at-limit */ + + if (avgidle < cl->minidle) + avgidle = cl->minidle; + + cl->avgidle = avgidle; + + /* Calculate expected time, when this class + * will be allowed to send. + * It will occur, when: + * (1-W)*true_avgidle + W*delay = 0, i.e. + * idle = (1/W - 1)*(-true_avgidle) + * or + * idle = (1 - W)*(-cl->avgidle); + */ + idle = (-avgidle) - ((-avgidle) >> cl->ewma_log); + + /* + * That is not all. + * To maintain the rate allocated to the class, + * we add to undertime virtual clock, + * necessary to complete transmitted packet. + * (len/phys_bandwidth has been already passed + * to the moment of cbq_update) + */ + + idle -= L2T(&q->link, len); + idle += L2T(cl, len); + + cl->undertime = now + idle; + } else { + /* Underlimit */ + + cl->undertime = PSCHED_PASTPERFECT; + if (avgidle > cl->maxidle) + cl->avgidle = cl->maxidle; + else + cl->avgidle = avgidle; + } + if ((s64)(now - cl->last) > 0) + cl->last = now; + } + + cbq_update_toplevel(q, this, q->tx_borrowed); +} + +static inline struct cbq_class * +cbq_under_limit(struct cbq_class *cl) +{ + struct cbq_sched_data *q = qdisc_priv(cl->qdisc); + struct cbq_class *this_cl = cl; + + if (cl->tparent == NULL) + return cl; + + if (cl->undertime == PSCHED_PASTPERFECT || q->now >= cl->undertime) { + cl->delayed = 0; + return cl; + } + + do { + /* It is very suspicious place. Now overlimit + * action is generated for not bounded classes + * only if link is completely congested. + * Though it is in agree with ancestor-only paradigm, + * it looks very stupid. Particularly, + * it means that this chunk of code will either + * never be called or result in strong amplification + * of burstiness. Dangerous, silly, and, however, + * no another solution exists. + */ + cl = cl->borrow; + if (!cl) { + this_cl->qstats.overlimits++; + cbq_overlimit(this_cl); + return NULL; + } + if (cl->level > q->toplevel) + return NULL; + } while (cl->undertime != PSCHED_PASTPERFECT && q->now < cl->undertime); + + cl->delayed = 0; + return cl; +} + +static inline struct sk_buff * +cbq_dequeue_prio(struct Qdisc *sch, int prio) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl_tail, *cl_prev, *cl; + struct sk_buff *skb; + int deficit; + + cl_tail = cl_prev = q->active[prio]; + cl = cl_prev->next_alive; + + do { + deficit = 0; + + /* Start round */ + do { + struct cbq_class *borrow = cl; + + if (cl->q->q.qlen && + (borrow = cbq_under_limit(cl)) == NULL) + goto skip_class; + + if (cl->deficit <= 0) { + /* Class exhausted its allotment per + * this round. Switch to the next one. + */ + deficit = 1; + cl->deficit += cl->quantum; + goto next_class; + } + + skb = cl->q->dequeue(cl->q); + + /* Class did not give us any skb :-( + * It could occur even if cl->q->q.qlen != 0 + * f.e. if cl->q == "tbf" + */ + if (skb == NULL) + goto skip_class; + + cl->deficit -= qdisc_pkt_len(skb); + q->tx_class = cl; + q->tx_borrowed = borrow; + if (borrow != cl) { +#ifndef CBQ_XSTATS_BORROWS_BYTES + borrow->xstats.borrows++; + cl->xstats.borrows++; +#else + borrow->xstats.borrows += qdisc_pkt_len(skb); + cl->xstats.borrows += qdisc_pkt_len(skb); +#endif + } + q->tx_len = qdisc_pkt_len(skb); + + if (cl->deficit <= 0) { + q->active[prio] = cl; + cl = cl->next_alive; + cl->deficit += cl->quantum; + } + return skb; + +skip_class: + if (cl->q->q.qlen == 0 || prio != cl->cpriority) { + /* Class is empty or penalized. + * Unlink it from active chain. + */ + cl_prev->next_alive = cl->next_alive; + cl->next_alive = NULL; + + /* Did cl_tail point to it? */ + if (cl == cl_tail) { + /* Repair it! */ + cl_tail = cl_prev; + + /* Was it the last class in this band? */ + if (cl == cl_tail) { + /* Kill the band! */ + q->active[prio] = NULL; + q->activemask &= ~(1<<prio); + if (cl->q->q.qlen) + cbq_activate_class(cl); + return NULL; + } + + q->active[prio] = cl_tail; + } + if (cl->q->q.qlen) + cbq_activate_class(cl); + + cl = cl_prev; + } + +next_class: + cl_prev = cl; + cl = cl->next_alive; + } while (cl_prev != cl_tail); + } while (deficit); + + q->active[prio] = cl_prev; + + return NULL; +} + +static inline struct sk_buff * +cbq_dequeue_1(struct Qdisc *sch) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + unsigned int activemask; + + activemask = q->activemask & 0xFF; + while (activemask) { + int prio = ffz(~activemask); + activemask &= ~(1<<prio); + skb = cbq_dequeue_prio(sch, prio); + if (skb) + return skb; + } + return NULL; +} + +static struct sk_buff * +cbq_dequeue(struct Qdisc *sch) +{ + struct sk_buff *skb; + struct cbq_sched_data *q = qdisc_priv(sch); + psched_time_t now; + + now = psched_get_time(); + + if (q->tx_class) + cbq_update(q); + + q->now = now; + + for (;;) { + q->wd_expires = 0; + + skb = cbq_dequeue_1(sch); + if (skb) { + qdisc_bstats_update(sch, skb); + sch->q.qlen--; + return skb; + } + + /* All the classes are overlimit. + * + * It is possible, if: + * + * 1. Scheduler is empty. + * 2. Toplevel cutoff inhibited borrowing. + * 3. Root class is overlimit. + * + * Reset 2d and 3d conditions and retry. + * + * Note, that NS and cbq-2.0 are buggy, peeking + * an arbitrary class is appropriate for ancestor-only + * sharing, but not for toplevel algorithm. + * + * Our version is better, but slower, because it requires + * two passes, but it is unavoidable with top-level sharing. + */ + + if (q->toplevel == TC_CBQ_MAXLEVEL && + q->link.undertime == PSCHED_PASTPERFECT) + break; + + q->toplevel = TC_CBQ_MAXLEVEL; + q->link.undertime = PSCHED_PASTPERFECT; + } + + /* No packets in scheduler or nobody wants to give them to us :-( + * Sigh... start watchdog timer in the last case. + */ + + if (sch->q.qlen) { + qdisc_qstats_overlimit(sch); + if (q->wd_expires) + qdisc_watchdog_schedule(&q->watchdog, + now + q->wd_expires); + } + return NULL; +} + +/* CBQ class maintanance routines */ + +static void cbq_adjust_levels(struct cbq_class *this) +{ + if (this == NULL) + return; + + do { + int level = 0; + struct cbq_class *cl; + + cl = this->children; + if (cl) { + do { + if (cl->level > level) + level = cl->level; + } while ((cl = cl->sibling) != this->children); + } + this->level = level + 1; + } while ((this = this->tparent) != NULL); +} + +static void cbq_normalize_quanta(struct cbq_sched_data *q, int prio) +{ + struct cbq_class *cl; + unsigned int h; + + if (q->quanta[prio] == 0) + return; + + for (h = 0; h < q->clhash.hashsize; h++) { + hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { + /* BUGGGG... Beware! This expression suffer of + * arithmetic overflows! + */ + if (cl->priority == prio) { + cl->quantum = (cl->weight*cl->allot*q->nclasses[prio])/ + q->quanta[prio]; + } + if (cl->quantum <= 0 || + cl->quantum > 32*qdisc_dev(cl->qdisc)->mtu) { + pr_warn("CBQ: class %08x has bad quantum==%ld, repaired.\n", + cl->common.classid, cl->quantum); + cl->quantum = qdisc_dev(cl->qdisc)->mtu/2 + 1; + } + } + } +} + +static void cbq_sync_defmap(struct cbq_class *cl) +{ + struct cbq_sched_data *q = qdisc_priv(cl->qdisc); + struct cbq_class *split = cl->split; + unsigned int h; + int i; + + if (split == NULL) + return; + + for (i = 0; i <= TC_PRIO_MAX; i++) { + if (split->defaults[i] == cl && !(cl->defmap & (1<<i))) + split->defaults[i] = NULL; + } + + for (i = 0; i <= TC_PRIO_MAX; i++) { + int level = split->level; + + if (split->defaults[i]) + continue; + + for (h = 0; h < q->clhash.hashsize; h++) { + struct cbq_class *c; + + hlist_for_each_entry(c, &q->clhash.hash[h], + common.hnode) { + if (c->split == split && c->level < level && + c->defmap & (1<<i)) { + split->defaults[i] = c; + level = c->level; + } + } + } + } +} + +static void cbq_change_defmap(struct cbq_class *cl, u32 splitid, u32 def, u32 mask) +{ + struct cbq_class *split = NULL; + + if (splitid == 0) { + split = cl->split; + if (!split) + return; + splitid = split->common.classid; + } + + if (split == NULL || split->common.classid != splitid) { + for (split = cl->tparent; split; split = split->tparent) + if (split->common.classid == splitid) + break; + } + + if (split == NULL) + return; + + if (cl->split != split) { + cl->defmap = 0; + cbq_sync_defmap(cl); + cl->split = split; + cl->defmap = def & mask; + } else + cl->defmap = (cl->defmap & ~mask) | (def & mask); + + cbq_sync_defmap(cl); +} + +static void cbq_unlink_class(struct cbq_class *this) +{ + struct cbq_class *cl, **clp; + struct cbq_sched_data *q = qdisc_priv(this->qdisc); + + qdisc_class_hash_remove(&q->clhash, &this->common); + + if (this->tparent) { + clp = &this->sibling; + cl = *clp; + do { + if (cl == this) { + *clp = cl->sibling; + break; + } + clp = &cl->sibling; + } while ((cl = *clp) != this->sibling); + + if (this->tparent->children == this) { + this->tparent->children = this->sibling; + if (this->sibling == this) + this->tparent->children = NULL; + } + } else { + WARN_ON(this->sibling != this); + } +} + +static void cbq_link_class(struct cbq_class *this) +{ + struct cbq_sched_data *q = qdisc_priv(this->qdisc); + struct cbq_class *parent = this->tparent; + + this->sibling = this; + qdisc_class_hash_insert(&q->clhash, &this->common); + + if (parent == NULL) + return; + + if (parent->children == NULL) { + parent->children = this; + } else { + this->sibling = parent->children->sibling; + parent->children->sibling = this; + } +} + +static void +cbq_reset(struct Qdisc *sch) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl; + int prio; + unsigned int h; + + q->activemask = 0; + q->pmask = 0; + q->tx_class = NULL; + q->tx_borrowed = NULL; + qdisc_watchdog_cancel(&q->watchdog); + hrtimer_cancel(&q->delay_timer); + q->toplevel = TC_CBQ_MAXLEVEL; + q->now = psched_get_time(); + + for (prio = 0; prio <= TC_CBQ_MAXPRIO; prio++) + q->active[prio] = NULL; + + for (h = 0; h < q->clhash.hashsize; h++) { + hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { + qdisc_reset(cl->q); + + cl->next_alive = NULL; + cl->undertime = PSCHED_PASTPERFECT; + cl->avgidle = cl->maxidle; + cl->deficit = cl->quantum; + cl->cpriority = cl->priority; + } + } + sch->q.qlen = 0; +} + + +static int cbq_set_lss(struct cbq_class *cl, struct tc_cbq_lssopt *lss) +{ + if (lss->change & TCF_CBQ_LSS_FLAGS) { + cl->share = (lss->flags & TCF_CBQ_LSS_ISOLATED) ? NULL : cl->tparent; + cl->borrow = (lss->flags & TCF_CBQ_LSS_BOUNDED) ? NULL : cl->tparent; + } + if (lss->change & TCF_CBQ_LSS_EWMA) + cl->ewma_log = lss->ewma_log; + if (lss->change & TCF_CBQ_LSS_AVPKT) + cl->avpkt = lss->avpkt; + if (lss->change & TCF_CBQ_LSS_MINIDLE) + cl->minidle = -(long)lss->minidle; + if (lss->change & TCF_CBQ_LSS_MAXIDLE) { + cl->maxidle = lss->maxidle; + cl->avgidle = lss->maxidle; + } + if (lss->change & TCF_CBQ_LSS_OFFTIME) + cl->offtime = lss->offtime; + return 0; +} + +static void cbq_rmprio(struct cbq_sched_data *q, struct cbq_class *cl) +{ + q->nclasses[cl->priority]--; + q->quanta[cl->priority] -= cl->weight; + cbq_normalize_quanta(q, cl->priority); +} + +static void cbq_addprio(struct cbq_sched_data *q, struct cbq_class *cl) +{ + q->nclasses[cl->priority]++; + q->quanta[cl->priority] += cl->weight; + cbq_normalize_quanta(q, cl->priority); +} + +static int cbq_set_wrr(struct cbq_class *cl, struct tc_cbq_wrropt *wrr) +{ + struct cbq_sched_data *q = qdisc_priv(cl->qdisc); + + if (wrr->allot) + cl->allot = wrr->allot; + if (wrr->weight) + cl->weight = wrr->weight; + if (wrr->priority) { + cl->priority = wrr->priority - 1; + cl->cpriority = cl->priority; + if (cl->priority >= cl->priority2) + cl->priority2 = TC_CBQ_MAXPRIO - 1; + } + + cbq_addprio(q, cl); + return 0; +} + +static int cbq_set_fopt(struct cbq_class *cl, struct tc_cbq_fopt *fopt) +{ + cbq_change_defmap(cl, fopt->split, fopt->defmap, fopt->defchange); + return 0; +} + +static const struct nla_policy cbq_policy[TCA_CBQ_MAX + 1] = { + [TCA_CBQ_LSSOPT] = { .len = sizeof(struct tc_cbq_lssopt) }, + [TCA_CBQ_WRROPT] = { .len = sizeof(struct tc_cbq_wrropt) }, + [TCA_CBQ_FOPT] = { .len = sizeof(struct tc_cbq_fopt) }, + [TCA_CBQ_OVL_STRATEGY] = { .len = sizeof(struct tc_cbq_ovl) }, + [TCA_CBQ_RATE] = { .len = sizeof(struct tc_ratespec) }, + [TCA_CBQ_RTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, + [TCA_CBQ_POLICE] = { .len = sizeof(struct tc_cbq_police) }, +}; + +static int cbq_opt_parse(struct nlattr *tb[TCA_CBQ_MAX + 1], + struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + int err; + + if (!opt) { + NL_SET_ERR_MSG(extack, "CBQ options are required for this operation"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_CBQ_MAX, opt, cbq_policy, extack); + if (err < 0) + return err; + + if (tb[TCA_CBQ_WRROPT]) { + const struct tc_cbq_wrropt *wrr = nla_data(tb[TCA_CBQ_WRROPT]); + + if (wrr->priority > TC_CBQ_MAXPRIO) { + NL_SET_ERR_MSG(extack, "priority is bigger than TC_CBQ_MAXPRIO"); + err = -EINVAL; + } + } + return err; +} + +static int cbq_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_CBQ_MAX + 1]; + struct tc_ratespec *r; + int err; + + qdisc_watchdog_init(&q->watchdog, sch); + hrtimer_init(&q->delay_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); + q->delay_timer.function = cbq_undelay; + + err = cbq_opt_parse(tb, opt, extack); + if (err < 0) + return err; + + if (!tb[TCA_CBQ_RTAB] || !tb[TCA_CBQ_RATE]) { + NL_SET_ERR_MSG(extack, "Rate specification missing or incomplete"); + return -EINVAL; + } + + r = nla_data(tb[TCA_CBQ_RATE]); + + q->link.R_tab = qdisc_get_rtab(r, tb[TCA_CBQ_RTAB], extack); + if (!q->link.R_tab) + return -EINVAL; + + err = tcf_block_get(&q->link.block, &q->link.filter_list, sch, extack); + if (err) + goto put_rtab; + + err = qdisc_class_hash_init(&q->clhash); + if (err < 0) + goto put_block; + + q->link.sibling = &q->link; + q->link.common.classid = sch->handle; + q->link.qdisc = sch; + q->link.q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + sch->handle, NULL); + if (!q->link.q) + q->link.q = &noop_qdisc; + else + qdisc_hash_add(q->link.q, true); + + q->link.priority = TC_CBQ_MAXPRIO - 1; + q->link.priority2 = TC_CBQ_MAXPRIO - 1; + q->link.cpriority = TC_CBQ_MAXPRIO - 1; + q->link.allot = psched_mtu(qdisc_dev(sch)); + q->link.quantum = q->link.allot; + q->link.weight = q->link.R_tab->rate.rate; + + q->link.ewma_log = TC_CBQ_DEF_EWMA; + q->link.avpkt = q->link.allot/2; + q->link.minidle = -0x7FFFFFFF; + + q->toplevel = TC_CBQ_MAXLEVEL; + q->now = psched_get_time(); + + cbq_link_class(&q->link); + + if (tb[TCA_CBQ_LSSOPT]) + cbq_set_lss(&q->link, nla_data(tb[TCA_CBQ_LSSOPT])); + + cbq_addprio(q, &q->link); + return 0; + +put_block: + tcf_block_put(q->link.block); + +put_rtab: + qdisc_put_rtab(q->link.R_tab); + return err; +} + +static int cbq_dump_rate(struct sk_buff *skb, struct cbq_class *cl) +{ + unsigned char *b = skb_tail_pointer(skb); + + if (nla_put(skb, TCA_CBQ_RATE, sizeof(cl->R_tab->rate), &cl->R_tab->rate)) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int cbq_dump_lss(struct sk_buff *skb, struct cbq_class *cl) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tc_cbq_lssopt opt; + + opt.flags = 0; + if (cl->borrow == NULL) + opt.flags |= TCF_CBQ_LSS_BOUNDED; + if (cl->share == NULL) + opt.flags |= TCF_CBQ_LSS_ISOLATED; + opt.ewma_log = cl->ewma_log; + opt.level = cl->level; + opt.avpkt = cl->avpkt; + opt.maxidle = cl->maxidle; + opt.minidle = (u32)(-cl->minidle); + opt.offtime = cl->offtime; + opt.change = ~0; + if (nla_put(skb, TCA_CBQ_LSSOPT, sizeof(opt), &opt)) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int cbq_dump_wrr(struct sk_buff *skb, struct cbq_class *cl) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tc_cbq_wrropt opt; + + memset(&opt, 0, sizeof(opt)); + opt.flags = 0; + opt.allot = cl->allot; + opt.priority = cl->priority + 1; + opt.cpriority = cl->cpriority + 1; + opt.weight = cl->weight; + if (nla_put(skb, TCA_CBQ_WRROPT, sizeof(opt), &opt)) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int cbq_dump_fopt(struct sk_buff *skb, struct cbq_class *cl) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tc_cbq_fopt opt; + + if (cl->split || cl->defmap) { + opt.split = cl->split ? cl->split->common.classid : 0; + opt.defmap = cl->defmap; + opt.defchange = ~0; + if (nla_put(skb, TCA_CBQ_FOPT, sizeof(opt), &opt)) + goto nla_put_failure; + } + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int cbq_dump_attr(struct sk_buff *skb, struct cbq_class *cl) +{ + if (cbq_dump_lss(skb, cl) < 0 || + cbq_dump_rate(skb, cl) < 0 || + cbq_dump_wrr(skb, cl) < 0 || + cbq_dump_fopt(skb, cl) < 0) + return -1; + return 0; +} + +static int cbq_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + if (cbq_dump_attr(skb, &q->link) < 0) + goto nla_put_failure; + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int +cbq_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + + q->link.xstats.avgidle = q->link.avgidle; + return gnet_stats_copy_app(d, &q->link.xstats, sizeof(q->link.xstats)); +} + +static int +cbq_dump_class(struct Qdisc *sch, unsigned long arg, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct cbq_class *cl = (struct cbq_class *)arg; + struct nlattr *nest; + + if (cl->tparent) + tcm->tcm_parent = cl->tparent->common.classid; + else + tcm->tcm_parent = TC_H_ROOT; + tcm->tcm_handle = cl->common.classid; + tcm->tcm_info = cl->q->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + if (cbq_dump_attr(skb, cl) < 0) + goto nla_put_failure; + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int +cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg, + struct gnet_dump *d) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl = (struct cbq_class *)arg; + + cl->xstats.avgidle = cl->avgidle; + cl->xstats.undertime = 0; + + if (cl->undertime != PSCHED_PASTPERFECT) + cl->xstats.undertime = cl->undertime - q->now; + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || + gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || + gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->q->q.qlen) < 0) + return -1; + + return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats)); +} + +static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct cbq_class *cl = (struct cbq_class *)arg; + + if (new == NULL) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + cl->common.classid, extack); + if (new == NULL) + return -ENOBUFS; + } + + *old = qdisc_replace(sch, new, &cl->q); + return 0; +} + +static struct Qdisc *cbq_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct cbq_class *cl = (struct cbq_class *)arg; + + return cl->q; +} + +static void cbq_qlen_notify(struct Qdisc *sch, unsigned long arg) +{ + struct cbq_class *cl = (struct cbq_class *)arg; + + cbq_deactivate_class(cl); +} + +static unsigned long cbq_find(struct Qdisc *sch, u32 classid) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + + return (unsigned long)cbq_class_lookup(q, classid); +} + +static void cbq_destroy_class(struct Qdisc *sch, struct cbq_class *cl) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + + WARN_ON(cl->filters); + + tcf_block_put(cl->block); + qdisc_put(cl->q); + qdisc_put_rtab(cl->R_tab); + gen_kill_estimator(&cl->rate_est); + if (cl != &q->link) + kfree(cl); +} + +static void cbq_destroy(struct Qdisc *sch) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct hlist_node *next; + struct cbq_class *cl; + unsigned int h; + +#ifdef CONFIG_NET_CLS_ACT + q->rx_class = NULL; +#endif + /* + * Filters must be destroyed first because we don't destroy the + * classes from root to leafs which means that filters can still + * be bound to classes which have been destroyed already. --TGR '04 + */ + for (h = 0; h < q->clhash.hashsize; h++) { + hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { + tcf_block_put(cl->block); + cl->block = NULL; + } + } + for (h = 0; h < q->clhash.hashsize; h++) { + hlist_for_each_entry_safe(cl, next, &q->clhash.hash[h], + common.hnode) + cbq_destroy_class(sch, cl); + } + qdisc_class_hash_destroy(&q->clhash); +} + +static int +cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **tca, + unsigned long *arg, struct netlink_ext_ack *extack) +{ + int err; + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl = (struct cbq_class *)*arg; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_CBQ_MAX + 1]; + struct cbq_class *parent; + struct qdisc_rate_table *rtab = NULL; + + err = cbq_opt_parse(tb, opt, extack); + if (err < 0) + return err; + + if (tb[TCA_CBQ_OVL_STRATEGY] || tb[TCA_CBQ_POLICE]) { + NL_SET_ERR_MSG(extack, "Neither overlimit strategy nor policing attributes can be used for changing class params"); + return -EOPNOTSUPP; + } + + if (cl) { + /* Check parent */ + if (parentid) { + if (cl->tparent && + cl->tparent->common.classid != parentid) { + NL_SET_ERR_MSG(extack, "Invalid parent id"); + return -EINVAL; + } + if (!cl->tparent && parentid != TC_H_ROOT) { + NL_SET_ERR_MSG(extack, "Parent must be root"); + return -EINVAL; + } + } + + if (tb[TCA_CBQ_RATE]) { + rtab = qdisc_get_rtab(nla_data(tb[TCA_CBQ_RATE]), + tb[TCA_CBQ_RTAB], extack); + if (rtab == NULL) + return -EINVAL; + } + + if (tca[TCA_RATE]) { + err = gen_replace_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) { + NL_SET_ERR_MSG(extack, "Failed to replace specified rate estimator"); + qdisc_put_rtab(rtab); + return err; + } + } + + /* Change class parameters */ + sch_tree_lock(sch); + + if (cl->next_alive != NULL) + cbq_deactivate_class(cl); + + if (rtab) { + qdisc_put_rtab(cl->R_tab); + cl->R_tab = rtab; + } + + if (tb[TCA_CBQ_LSSOPT]) + cbq_set_lss(cl, nla_data(tb[TCA_CBQ_LSSOPT])); + + if (tb[TCA_CBQ_WRROPT]) { + cbq_rmprio(q, cl); + cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT])); + } + + if (tb[TCA_CBQ_FOPT]) + cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT])); + + if (cl->q->q.qlen) + cbq_activate_class(cl); + + sch_tree_unlock(sch); + + return 0; + } + + if (parentid == TC_H_ROOT) + return -EINVAL; + + if (!tb[TCA_CBQ_WRROPT] || !tb[TCA_CBQ_RATE] || !tb[TCA_CBQ_LSSOPT]) { + NL_SET_ERR_MSG(extack, "One of the following attributes MUST be specified: WRR, rate or link sharing"); + return -EINVAL; + } + + rtab = qdisc_get_rtab(nla_data(tb[TCA_CBQ_RATE]), tb[TCA_CBQ_RTAB], + extack); + if (rtab == NULL) + return -EINVAL; + + if (classid) { + err = -EINVAL; + if (TC_H_MAJ(classid ^ sch->handle) || + cbq_class_lookup(q, classid)) { + NL_SET_ERR_MSG(extack, "Specified class not found"); + goto failure; + } + } else { + int i; + classid = TC_H_MAKE(sch->handle, 0x8000); + + for (i = 0; i < 0x8000; i++) { + if (++q->hgenerator >= 0x8000) + q->hgenerator = 1; + if (cbq_class_lookup(q, classid|q->hgenerator) == NULL) + break; + } + err = -ENOSR; + if (i >= 0x8000) { + NL_SET_ERR_MSG(extack, "Unable to generate classid"); + goto failure; + } + classid = classid|q->hgenerator; + } + + parent = &q->link; + if (parentid) { + parent = cbq_class_lookup(q, parentid); + err = -EINVAL; + if (!parent) { + NL_SET_ERR_MSG(extack, "Failed to find parentid"); + goto failure; + } + } + + err = -ENOBUFS; + cl = kzalloc(sizeof(*cl), GFP_KERNEL); + if (cl == NULL) + goto failure; + + err = tcf_block_get(&cl->block, &cl->filter_list, sch, extack); + if (err) { + kfree(cl); + goto failure; + } + + if (tca[TCA_RATE]) { + err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) { + NL_SET_ERR_MSG(extack, "Couldn't create new estimator"); + tcf_block_put(cl->block); + kfree(cl); + goto failure; + } + } + + cl->R_tab = rtab; + rtab = NULL; + cl->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, classid, + NULL); + if (!cl->q) + cl->q = &noop_qdisc; + else + qdisc_hash_add(cl->q, true); + + cl->common.classid = classid; + cl->tparent = parent; + cl->qdisc = sch; + cl->allot = parent->allot; + cl->quantum = cl->allot; + cl->weight = cl->R_tab->rate.rate; + + sch_tree_lock(sch); + cbq_link_class(cl); + cl->borrow = cl->tparent; + if (cl->tparent != &q->link) + cl->share = cl->tparent; + cbq_adjust_levels(parent); + cl->minidle = -0x7FFFFFFF; + cbq_set_lss(cl, nla_data(tb[TCA_CBQ_LSSOPT])); + cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT])); + if (cl->ewma_log == 0) + cl->ewma_log = q->link.ewma_log; + if (cl->maxidle == 0) + cl->maxidle = q->link.maxidle; + if (cl->avpkt == 0) + cl->avpkt = q->link.avpkt; + if (tb[TCA_CBQ_FOPT]) + cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT])); + sch_tree_unlock(sch); + + qdisc_class_hash_grow(sch, &q->clhash); + + *arg = (unsigned long)cl; + return 0; + +failure: + qdisc_put_rtab(rtab); + return err; +} + +static int cbq_delete(struct Qdisc *sch, unsigned long arg) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl = (struct cbq_class *)arg; + unsigned int qlen, backlog; + + if (cl->filters || cl->children || cl == &q->link) + return -EBUSY; + + sch_tree_lock(sch); + + qlen = cl->q->q.qlen; + backlog = cl->q->qstats.backlog; + qdisc_reset(cl->q); + qdisc_tree_reduce_backlog(cl->q, qlen, backlog); + + if (cl->next_alive) + cbq_deactivate_class(cl); + + if (q->tx_borrowed == cl) + q->tx_borrowed = q->tx_class; + if (q->tx_class == cl) { + q->tx_class = NULL; + q->tx_borrowed = NULL; + } +#ifdef CONFIG_NET_CLS_ACT + if (q->rx_class == cl) + q->rx_class = NULL; +#endif + + cbq_unlink_class(cl); + cbq_adjust_levels(cl->tparent); + cl->defmap = 0; + cbq_sync_defmap(cl); + + cbq_rmprio(q, cl); + sch_tree_unlock(sch); + + cbq_destroy_class(sch, cl); + return 0; +} + +static struct tcf_block *cbq_tcf_block(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl = (struct cbq_class *)arg; + + if (cl == NULL) + cl = &q->link; + + return cl->block; +} + +static unsigned long cbq_bind_filter(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *p = (struct cbq_class *)parent; + struct cbq_class *cl = cbq_class_lookup(q, classid); + + if (cl) { + if (p && p->level <= cl->level) + return 0; + cl->filters++; + return (unsigned long)cl; + } + return 0; +} + +static void cbq_unbind_filter(struct Qdisc *sch, unsigned long arg) +{ + struct cbq_class *cl = (struct cbq_class *)arg; + + cl->filters--; +} + +static void cbq_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct cbq_sched_data *q = qdisc_priv(sch); + struct cbq_class *cl; + unsigned int h; + + if (arg->stop) + return; + + for (h = 0; h < q->clhash.hashsize; h++) { + hlist_for_each_entry(cl, &q->clhash.hash[h], common.hnode) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, (unsigned long)cl, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static const struct Qdisc_class_ops cbq_class_ops = { + .graft = cbq_graft, + .leaf = cbq_leaf, + .qlen_notify = cbq_qlen_notify, + .find = cbq_find, + .change = cbq_change_class, + .delete = cbq_delete, + .walk = cbq_walk, + .tcf_block = cbq_tcf_block, + .bind_tcf = cbq_bind_filter, + .unbind_tcf = cbq_unbind_filter, + .dump = cbq_dump_class, + .dump_stats = cbq_dump_class_stats, +}; + +static struct Qdisc_ops cbq_qdisc_ops __read_mostly = { + .next = NULL, + .cl_ops = &cbq_class_ops, + .id = "cbq", + .priv_size = sizeof(struct cbq_sched_data), + .enqueue = cbq_enqueue, + .dequeue = cbq_dequeue, + .peek = qdisc_peek_dequeued, + .init = cbq_init, + .reset = cbq_reset, + .destroy = cbq_destroy, + .change = NULL, + .dump = cbq_dump, + .dump_stats = cbq_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init cbq_module_init(void) +{ + return register_qdisc(&cbq_qdisc_ops); +} +static void __exit cbq_module_exit(void) +{ + unregister_qdisc(&cbq_qdisc_ops); +} +module_init(cbq_module_init) +module_exit(cbq_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c new file mode 100644 index 000000000..f95dc899e --- /dev/null +++ b/net/sched/sch_cbs.c @@ -0,0 +1,588 @@ +/* + * net/sched/sch_cbs.c Credit Based Shaper + * + * 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. + * + * Authors: Vinicius Costa Gomes <vinicius.gomes@intel.com> + * + */ + +/* Credit Based Shaper (CBS) + * ========================= + * + * This is a simple rate-limiting shaper aimed at TSN applications on + * systems with known traffic workloads. + * + * Its algorithm is defined by the IEEE 802.1Q-2014 Specification, + * Section 8.6.8.2, and explained in more detail in the Annex L of the + * same specification. + * + * There are four tunables to be considered: + * + * 'idleslope': Idleslope is the rate of credits that is + * accumulated (in kilobits per second) when there is at least + * one packet waiting for transmission. Packets are transmitted + * when the current value of credits is equal or greater than + * zero. When there is no packet to be transmitted the amount of + * credits is set to zero. This is the main tunable of the CBS + * algorithm. + * + * 'sendslope': + * Sendslope is the rate of credits that is depleted (it should be a + * negative number of kilobits per second) when a transmission is + * ocurring. It can be calculated as follows, (IEEE 802.1Q-2014 Section + * 8.6.8.2 item g): + * + * sendslope = idleslope - port_transmit_rate + * + * 'hicredit': Hicredit defines the maximum amount of credits (in + * bytes) that can be accumulated. Hicredit depends on the + * characteristics of interfering traffic, + * 'max_interference_size' is the maximum size of any burst of + * traffic that can delay the transmission of a frame that is + * available for transmission for this traffic class, (IEEE + * 802.1Q-2014 Annex L, Equation L-3): + * + * hicredit = max_interference_size * (idleslope / port_transmit_rate) + * + * 'locredit': Locredit is the minimum amount of credits that can + * be reached. It is a function of the traffic flowing through + * this qdisc (IEEE 802.1Q-2014 Annex L, Equation L-2): + * + * locredit = max_frame_size * (sendslope / port_transmit_rate) + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netevent.h> +#include <net/netlink.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> + +static LIST_HEAD(cbs_list); +static DEFINE_SPINLOCK(cbs_list_lock); + +#define BYTES_PER_KBIT (1000LL / 8) + +struct cbs_sched_data { + bool offload; + int queue; + atomic64_t port_rate; /* in bytes/s */ + s64 last; /* timestamp in ns */ + s64 credits; /* in bytes */ + s32 locredit; /* in bytes */ + s32 hicredit; /* in bytes */ + s64 sendslope; /* in bytes/s */ + s64 idleslope; /* in bytes/s */ + struct qdisc_watchdog watchdog; + int (*enqueue)(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free); + struct sk_buff *(*dequeue)(struct Qdisc *sch); + struct Qdisc *qdisc; + struct list_head cbs_list; +}; + +static int cbs_child_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct Qdisc *child, + struct sk_buff **to_free) +{ + int err; + + err = child->ops->enqueue(skb, child, to_free); + if (err != NET_XMIT_SUCCESS) + return err; + + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + + return NET_XMIT_SUCCESS; +} + +static int cbs_enqueue_offload(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct Qdisc *qdisc = q->qdisc; + + return cbs_child_enqueue(skb, sch, qdisc, to_free); +} + +static int cbs_enqueue_soft(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct Qdisc *qdisc = q->qdisc; + + if (sch->q.qlen == 0 && q->credits > 0) { + /* We need to stop accumulating credits when there's + * no enqueued packets and q->credits is positive. + */ + q->credits = 0; + q->last = ktime_get_ns(); + } + + return cbs_child_enqueue(skb, sch, qdisc, to_free); +} + +static int cbs_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + + return q->enqueue(skb, sch, to_free); +} + +/* timediff is in ns, slope is in bytes/s */ +static s64 timediff_to_credits(s64 timediff, s64 slope) +{ + return div64_s64(timediff * slope, NSEC_PER_SEC); +} + +static s64 delay_from_credits(s64 credits, s64 slope) +{ + if (unlikely(slope == 0)) + return S64_MAX; + + return div64_s64(-credits * NSEC_PER_SEC, slope); +} + +static s64 credits_from_len(unsigned int len, s64 slope, s64 port_rate) +{ + if (unlikely(port_rate == 0)) + return S64_MAX; + + return div64_s64(len * slope, port_rate); +} + +static struct sk_buff *cbs_child_dequeue(struct Qdisc *sch, struct Qdisc *child) +{ + struct sk_buff *skb; + + skb = child->ops->dequeue(child); + if (!skb) + return NULL; + + qdisc_qstats_backlog_dec(sch, skb); + qdisc_bstats_update(sch, skb); + sch->q.qlen--; + + return skb; +} + +static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct Qdisc *qdisc = q->qdisc; + s64 now = ktime_get_ns(); + struct sk_buff *skb; + s64 credits; + int len; + + /* The previous packet is still being sent */ + if (now < q->last) { + qdisc_watchdog_schedule_ns(&q->watchdog, q->last); + return NULL; + } + if (q->credits < 0) { + credits = timediff_to_credits(now - q->last, q->idleslope); + + credits = q->credits + credits; + q->credits = min_t(s64, credits, q->hicredit); + + if (q->credits < 0) { + s64 delay; + + delay = delay_from_credits(q->credits, q->idleslope); + qdisc_watchdog_schedule_ns(&q->watchdog, now + delay); + + q->last = now; + + return NULL; + } + } + skb = cbs_child_dequeue(sch, qdisc); + if (!skb) + return NULL; + + len = qdisc_pkt_len(skb); + + /* As sendslope is a negative number, this will decrease the + * amount of q->credits. + */ + credits = credits_from_len(len, q->sendslope, + atomic64_read(&q->port_rate)); + credits += q->credits; + + q->credits = max_t(s64, credits, q->locredit); + /* Estimate of the transmission of the last byte of the packet in ns */ + if (unlikely(atomic64_read(&q->port_rate) == 0)) + q->last = now; + else + q->last = now + div64_s64(len * NSEC_PER_SEC, + atomic64_read(&q->port_rate)); + + return skb; +} + +static struct sk_buff *cbs_dequeue_offload(struct Qdisc *sch) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct Qdisc *qdisc = q->qdisc; + + return cbs_child_dequeue(sch, qdisc); +} + +static struct sk_buff *cbs_dequeue(struct Qdisc *sch) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + + return q->dequeue(sch); +} + +static const struct nla_policy cbs_policy[TCA_CBS_MAX + 1] = { + [TCA_CBS_PARMS] = { .len = sizeof(struct tc_cbs_qopt) }, +}; + +static void cbs_disable_offload(struct net_device *dev, + struct cbs_sched_data *q) +{ + struct tc_cbs_qopt_offload cbs = { }; + const struct net_device_ops *ops; + int err; + + if (!q->offload) + return; + + q->enqueue = cbs_enqueue_soft; + q->dequeue = cbs_dequeue_soft; + + ops = dev->netdev_ops; + if (!ops->ndo_setup_tc) + return; + + cbs.queue = q->queue; + cbs.enable = 0; + + err = ops->ndo_setup_tc(dev, TC_SETUP_QDISC_CBS, &cbs); + if (err < 0) + pr_warn("Couldn't disable CBS offload for queue %d\n", + cbs.queue); +} + +static int cbs_enable_offload(struct net_device *dev, struct cbs_sched_data *q, + const struct tc_cbs_qopt *opt, + struct netlink_ext_ack *extack) +{ + const struct net_device_ops *ops = dev->netdev_ops; + struct tc_cbs_qopt_offload cbs = { }; + int err; + + if (!ops->ndo_setup_tc) { + NL_SET_ERR_MSG(extack, "Specified device does not support cbs offload"); + return -EOPNOTSUPP; + } + + cbs.queue = q->queue; + + cbs.enable = 1; + cbs.hicredit = opt->hicredit; + cbs.locredit = opt->locredit; + cbs.idleslope = opt->idleslope; + cbs.sendslope = opt->sendslope; + + err = ops->ndo_setup_tc(dev, TC_SETUP_QDISC_CBS, &cbs); + if (err < 0) { + NL_SET_ERR_MSG(extack, "Specified device failed to setup cbs hardware offload"); + return err; + } + + q->enqueue = cbs_enqueue_offload; + q->dequeue = cbs_dequeue_offload; + + return 0; +} + +static void cbs_set_port_rate(struct net_device *dev, struct cbs_sched_data *q) +{ + struct ethtool_link_ksettings ecmd; + int speed = SPEED_10; + int port_rate = -1; + int err; + + err = __ethtool_get_link_ksettings(dev, &ecmd); + if (err < 0) + goto skip; + + if (ecmd.base.speed && ecmd.base.speed != SPEED_UNKNOWN) + speed = ecmd.base.speed; + +skip: + port_rate = speed * 1000 * BYTES_PER_KBIT; + + atomic64_set(&q->port_rate, port_rate); + netdev_dbg(dev, "cbs: set %s's port_rate to: %lld, linkspeed: %d\n", + dev->name, (long long)atomic64_read(&q->port_rate), + ecmd.base.speed); +} + +static int cbs_dev_notifier(struct notifier_block *nb, unsigned long event, + void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct cbs_sched_data *q; + struct net_device *qdev; + bool found = false; + + ASSERT_RTNL(); + + if (event != NETDEV_UP && event != NETDEV_CHANGE) + return NOTIFY_DONE; + + spin_lock(&cbs_list_lock); + list_for_each_entry(q, &cbs_list, cbs_list) { + qdev = qdisc_dev(q->qdisc); + if (qdev == dev) { + found = true; + break; + } + } + spin_unlock(&cbs_list_lock); + + if (found) + cbs_set_port_rate(dev, q); + + return NOTIFY_DONE; +} + +static int cbs_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + struct nlattr *tb[TCA_CBS_MAX + 1]; + struct tc_cbs_qopt *qopt; + int err; + + err = nla_parse_nested(tb, TCA_CBS_MAX, opt, cbs_policy, extack); + if (err < 0) + return err; + + if (!tb[TCA_CBS_PARMS]) { + NL_SET_ERR_MSG(extack, "Missing CBS parameter which are mandatory"); + return -EINVAL; + } + + qopt = nla_data(tb[TCA_CBS_PARMS]); + + if (!qopt->offload) { + cbs_set_port_rate(dev, q); + cbs_disable_offload(dev, q); + } else { + err = cbs_enable_offload(dev, q, qopt, extack); + if (err < 0) + return err; + } + + /* Everything went OK, save the parameters used. */ + q->hicredit = qopt->hicredit; + q->locredit = qopt->locredit; + q->idleslope = qopt->idleslope * BYTES_PER_KBIT; + q->sendslope = qopt->sendslope * BYTES_PER_KBIT; + q->offload = qopt->offload; + + return 0; +} + +static int cbs_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + int err; + + if (!opt) { + NL_SET_ERR_MSG(extack, "Missing CBS qdisc options which are mandatory"); + return -EINVAL; + } + + q->qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + sch->handle, extack); + if (!q->qdisc) + return -ENOMEM; + + qdisc_hash_add(q->qdisc, false); + + q->queue = sch->dev_queue - netdev_get_tx_queue(dev, 0); + + q->enqueue = cbs_enqueue_soft; + q->dequeue = cbs_dequeue_soft; + + qdisc_watchdog_init(&q->watchdog, sch); + + err = cbs_change(sch, opt, extack); + if (err) + return err; + + if (!q->offload) { + spin_lock(&cbs_list_lock); + list_add(&q->cbs_list, &cbs_list); + spin_unlock(&cbs_list_lock); + } + + return 0; +} + +static void cbs_destroy(struct Qdisc *sch) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + + spin_lock(&cbs_list_lock); + list_del(&q->cbs_list); + spin_unlock(&cbs_list_lock); + + qdisc_watchdog_cancel(&q->watchdog); + cbs_disable_offload(dev, q); + + if (q->qdisc) + qdisc_put(q->qdisc); +} + +static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + struct tc_cbs_qopt opt = { }; + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + opt.hicredit = q->hicredit; + opt.locredit = q->locredit; + opt.sendslope = div64_s64(q->sendslope, BYTES_PER_KBIT); + opt.idleslope = div64_s64(q->idleslope, BYTES_PER_KBIT); + opt.offload = q->offload; + + if (nla_put(skb, TCA_CBS_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int cbs_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + + if (cl != 1 || !q->qdisc) /* only one class */ + return -ENOENT; + + tcm->tcm_handle |= TC_H_MIN(1); + tcm->tcm_info = q->qdisc->handle; + + return 0; +} + +static int cbs_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + + if (!new) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + sch->handle, NULL); + if (!new) + new = &noop_qdisc; + } + + *old = qdisc_replace(sch, new, &q->qdisc); + return 0; +} + +static struct Qdisc *cbs_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct cbs_sched_data *q = qdisc_priv(sch); + + return q->qdisc; +} + +static unsigned long cbs_find(struct Qdisc *sch, u32 classid) +{ + return 1; +} + +static void cbs_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + if (!walker->stop) { + if (walker->count >= walker->skip) { + if (walker->fn(sch, 1, walker) < 0) { + walker->stop = 1; + return; + } + } + walker->count++; + } +} + +static const struct Qdisc_class_ops cbs_class_ops = { + .graft = cbs_graft, + .leaf = cbs_leaf, + .find = cbs_find, + .walk = cbs_walk, + .dump = cbs_dump_class, +}; + +static struct Qdisc_ops cbs_qdisc_ops __read_mostly = { + .id = "cbs", + .cl_ops = &cbs_class_ops, + .priv_size = sizeof(struct cbs_sched_data), + .enqueue = cbs_enqueue, + .dequeue = cbs_dequeue, + .peek = qdisc_peek_dequeued, + .init = cbs_init, + .reset = qdisc_reset_queue, + .destroy = cbs_destroy, + .change = cbs_change, + .dump = cbs_dump, + .owner = THIS_MODULE, +}; + +static struct notifier_block cbs_device_notifier = { + .notifier_call = cbs_dev_notifier, +}; + +static int __init cbs_module_init(void) +{ + int err; + + err = register_netdevice_notifier(&cbs_device_notifier); + if (err) + return err; + + err = register_qdisc(&cbs_qdisc_ops); + if (err) + unregister_netdevice_notifier(&cbs_device_notifier); + + return err; +} + +static void __exit cbs_module_exit(void) +{ + unregister_qdisc(&cbs_qdisc_ops); + unregister_netdevice_notifier(&cbs_device_notifier); +} +module_init(cbs_module_init) +module_exit(cbs_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c new file mode 100644 index 000000000..8b72aed77 --- /dev/null +++ b/net/sched/sch_choke.c @@ -0,0 +1,528 @@ +/* + * net/sched/sch_choke.c CHOKE scheduler + * + * Copyright (c) 2011 Stephen Hemminger <shemminger@vyatta.com> + * Copyright (c) 2011 Eric Dumazet <eric.dumazet@gmail.com> + * + * 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/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <linux/vmalloc.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/inet_ecn.h> +#include <net/red.h> +#include <net/flow_dissector.h> + +/* + CHOKe stateless AQM for fair bandwidth allocation + ================================================= + + CHOKe (CHOose and Keep for responsive flows, CHOose and Kill for + unresponsive flows) is a variant of RED that penalizes misbehaving flows but + maintains no flow state. The difference from RED is an additional step + during the enqueuing process. If average queue size is over the + low threshold (qmin), a packet is chosen at random from the queue. + If both the new and chosen packet are from the same flow, both + are dropped. Unlike RED, CHOKe is not really a "classful" qdisc because it + needs to access packets in queue randomly. It has a minimal class + interface to allow overriding the builtin flow classifier with + filters. + + Source: + R. Pan, B. Prabhakar, and K. Psounis, "CHOKe, A Stateless + Active Queue Management Scheme for Approximating Fair Bandwidth Allocation", + IEEE INFOCOM, 2000. + + A. Tang, J. Wang, S. Low, "Understanding CHOKe: Throughput and Spatial + Characteristics", IEEE/ACM Transactions on Networking, 2004 + + */ + +/* Upper bound on size of sk_buff table (packets) */ +#define CHOKE_MAX_QUEUE (128*1024 - 1) + +struct choke_sched_data { +/* Parameters */ + u32 limit; + unsigned char flags; + + struct red_parms parms; + +/* Variables */ + struct red_vars vars; + struct { + u32 prob_drop; /* Early probability drops */ + u32 prob_mark; /* Early probability marks */ + u32 forced_drop; /* Forced drops, qavg > max_thresh */ + u32 forced_mark; /* Forced marks, qavg > max_thresh */ + u32 pdrop; /* Drops due to queue limits */ + u32 other; /* Drops due to drop() calls */ + u32 matched; /* Drops to flow match */ + } stats; + + unsigned int head; + unsigned int tail; + + unsigned int tab_mask; /* size - 1 */ + + struct sk_buff **tab; +}; + +/* number of elements in queue including holes */ +static unsigned int choke_len(const struct choke_sched_data *q) +{ + return (q->tail - q->head) & q->tab_mask; +} + +/* Is ECN parameter configured */ +static int use_ecn(const struct choke_sched_data *q) +{ + return q->flags & TC_RED_ECN; +} + +/* Should packets over max just be dropped (versus marked) */ +static int use_harddrop(const struct choke_sched_data *q) +{ + return q->flags & TC_RED_HARDDROP; +} + +/* Move head pointer forward to skip over holes */ +static void choke_zap_head_holes(struct choke_sched_data *q) +{ + do { + q->head = (q->head + 1) & q->tab_mask; + if (q->head == q->tail) + break; + } while (q->tab[q->head] == NULL); +} + +/* Move tail pointer backwards to reuse holes */ +static void choke_zap_tail_holes(struct choke_sched_data *q) +{ + do { + q->tail = (q->tail - 1) & q->tab_mask; + if (q->head == q->tail) + break; + } while (q->tab[q->tail] == NULL); +} + +/* Drop packet from queue array by creating a "hole" */ +static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx, + struct sk_buff **to_free) +{ + struct choke_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb = q->tab[idx]; + + q->tab[idx] = NULL; + + if (idx == q->head) + choke_zap_head_holes(q); + if (idx == q->tail) + choke_zap_tail_holes(q); + + qdisc_qstats_backlog_dec(sch, skb); + qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb)); + qdisc_drop(skb, sch, to_free); + --sch->q.qlen; +} + +struct choke_skb_cb { + u16 classid; + u8 keys_valid; + struct flow_keys_digest keys; +}; + +static inline struct choke_skb_cb *choke_skb_cb(const struct sk_buff *skb) +{ + qdisc_cb_private_validate(skb, sizeof(struct choke_skb_cb)); + return (struct choke_skb_cb *)qdisc_skb_cb(skb)->data; +} + +static inline void choke_set_classid(struct sk_buff *skb, u16 classid) +{ + choke_skb_cb(skb)->classid = classid; +} + +/* + * Compare flow of two packets + * Returns true only if source and destination address and port match. + * false for special cases + */ +static bool choke_match_flow(struct sk_buff *skb1, + struct sk_buff *skb2) +{ + struct flow_keys temp; + + if (skb1->protocol != skb2->protocol) + return false; + + if (!choke_skb_cb(skb1)->keys_valid) { + choke_skb_cb(skb1)->keys_valid = 1; + skb_flow_dissect_flow_keys(skb1, &temp, 0); + make_flow_keys_digest(&choke_skb_cb(skb1)->keys, &temp); + } + + if (!choke_skb_cb(skb2)->keys_valid) { + choke_skb_cb(skb2)->keys_valid = 1; + skb_flow_dissect_flow_keys(skb2, &temp, 0); + make_flow_keys_digest(&choke_skb_cb(skb2)->keys, &temp); + } + + return !memcmp(&choke_skb_cb(skb1)->keys, + &choke_skb_cb(skb2)->keys, + sizeof(choke_skb_cb(skb1)->keys)); +} + +/* + * Select a packet at random from queue + * HACK: since queue can have holes from previous deletion; retry several + * times to find a random skb but then just give up and return the head + * Will return NULL if queue is empty (q->head == q->tail) + */ +static struct sk_buff *choke_peek_random(const struct choke_sched_data *q, + unsigned int *pidx) +{ + struct sk_buff *skb; + int retrys = 3; + + do { + *pidx = (q->head + prandom_u32_max(choke_len(q))) & q->tab_mask; + skb = q->tab[*pidx]; + if (skb) + return skb; + } while (--retrys > 0); + + return q->tab[*pidx = q->head]; +} + +/* + * Compare new packet with random packet in queue + * returns true if matched and sets *pidx + */ +static bool choke_match_random(const struct choke_sched_data *q, + struct sk_buff *nskb, + unsigned int *pidx) +{ + struct sk_buff *oskb; + + if (q->head == q->tail) + return false; + + oskb = choke_peek_random(q, pidx); + return choke_match_flow(oskb, nskb); +} + +static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct choke_sched_data *q = qdisc_priv(sch); + const struct red_parms *p = &q->parms; + + choke_skb_cb(skb)->keys_valid = 0; + /* Compute average queue usage (see RED) */ + q->vars.qavg = red_calc_qavg(p, &q->vars, sch->q.qlen); + if (red_is_idling(&q->vars)) + red_end_of_idle_period(&q->vars); + + /* Is queue small? */ + if (q->vars.qavg <= p->qth_min) + q->vars.qcount = -1; + else { + unsigned int idx; + + /* Draw a packet at random from queue and compare flow */ + if (choke_match_random(q, skb, &idx)) { + q->stats.matched++; + choke_drop_by_idx(sch, idx, to_free); + goto congestion_drop; + } + + /* Queue is large, always mark/drop */ + if (q->vars.qavg > p->qth_max) { + q->vars.qcount = -1; + + qdisc_qstats_overlimit(sch); + if (use_harddrop(q) || !use_ecn(q) || + !INET_ECN_set_ce(skb)) { + q->stats.forced_drop++; + goto congestion_drop; + } + + q->stats.forced_mark++; + } else if (++q->vars.qcount) { + if (red_mark_probability(p, &q->vars, q->vars.qavg)) { + q->vars.qcount = 0; + q->vars.qR = red_random(p); + + qdisc_qstats_overlimit(sch); + if (!use_ecn(q) || !INET_ECN_set_ce(skb)) { + q->stats.prob_drop++; + goto congestion_drop; + } + + q->stats.prob_mark++; + } + } else + q->vars.qR = red_random(p); + } + + /* Admit new packet */ + if (sch->q.qlen < q->limit) { + q->tab[q->tail] = skb; + q->tail = (q->tail + 1) & q->tab_mask; + ++sch->q.qlen; + qdisc_qstats_backlog_inc(sch, skb); + return NET_XMIT_SUCCESS; + } + + q->stats.pdrop++; + return qdisc_drop(skb, sch, to_free); + +congestion_drop: + qdisc_drop(skb, sch, to_free); + return NET_XMIT_CN; +} + +static struct sk_buff *choke_dequeue(struct Qdisc *sch) +{ + struct choke_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + + if (q->head == q->tail) { + if (!red_is_idling(&q->vars)) + red_start_of_idle_period(&q->vars); + return NULL; + } + + skb = q->tab[q->head]; + q->tab[q->head] = NULL; + choke_zap_head_holes(q); + --sch->q.qlen; + qdisc_qstats_backlog_dec(sch, skb); + qdisc_bstats_update(sch, skb); + + return skb; +} + +static void choke_reset(struct Qdisc *sch) +{ + struct choke_sched_data *q = qdisc_priv(sch); + + while (q->head != q->tail) { + struct sk_buff *skb = q->tab[q->head]; + + q->head = (q->head + 1) & q->tab_mask; + if (!skb) + continue; + rtnl_qdisc_drop(skb, sch); + } + + sch->q.qlen = 0; + sch->qstats.backlog = 0; + if (q->tab) + memset(q->tab, 0, (q->tab_mask + 1) * sizeof(struct sk_buff *)); + q->head = q->tail = 0; + red_restart(&q->vars); +} + +static const struct nla_policy choke_policy[TCA_CHOKE_MAX + 1] = { + [TCA_CHOKE_PARMS] = { .len = sizeof(struct tc_red_qopt) }, + [TCA_CHOKE_STAB] = { .len = RED_STAB_SIZE }, + [TCA_CHOKE_MAX_P] = { .type = NLA_U32 }, +}; + + +static void choke_free(void *addr) +{ + kvfree(addr); +} + +static int choke_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct choke_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_CHOKE_MAX + 1]; + const struct tc_red_qopt *ctl; + int err; + struct sk_buff **old = NULL; + unsigned int mask; + u32 max_P; + u8 *stab; + + if (opt == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_CHOKE_MAX, opt, choke_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_CHOKE_PARMS] == NULL || + tb[TCA_CHOKE_STAB] == NULL) + return -EINVAL; + + max_P = tb[TCA_CHOKE_MAX_P] ? nla_get_u32(tb[TCA_CHOKE_MAX_P]) : 0; + + ctl = nla_data(tb[TCA_CHOKE_PARMS]); + stab = nla_data(tb[TCA_CHOKE_STAB]); + if (!red_check_params(ctl->qth_min, ctl->qth_max, ctl->Wlog, ctl->Scell_log, stab)) + return -EINVAL; + + if (ctl->limit > CHOKE_MAX_QUEUE) + return -EINVAL; + + mask = roundup_pow_of_two(ctl->limit + 1) - 1; + if (mask != q->tab_mask) { + struct sk_buff **ntab; + + ntab = kvmalloc_array((mask + 1), sizeof(struct sk_buff *), GFP_KERNEL | __GFP_ZERO); + if (!ntab) + return -ENOMEM; + + sch_tree_lock(sch); + old = q->tab; + if (old) { + unsigned int oqlen = sch->q.qlen, tail = 0; + unsigned dropped = 0; + + while (q->head != q->tail) { + struct sk_buff *skb = q->tab[q->head]; + + q->head = (q->head + 1) & q->tab_mask; + if (!skb) + continue; + if (tail < mask) { + ntab[tail++] = skb; + continue; + } + dropped += qdisc_pkt_len(skb); + qdisc_qstats_backlog_dec(sch, skb); + --sch->q.qlen; + rtnl_qdisc_drop(skb, sch); + } + qdisc_tree_reduce_backlog(sch, oqlen - sch->q.qlen, dropped); + q->head = 0; + q->tail = tail; + } + + q->tab_mask = mask; + q->tab = ntab; + } else + sch_tree_lock(sch); + + q->flags = ctl->flags; + q->limit = ctl->limit; + + red_set_parms(&q->parms, ctl->qth_min, ctl->qth_max, ctl->Wlog, + ctl->Plog, ctl->Scell_log, + stab, + max_P); + red_set_vars(&q->vars); + + if (q->head == q->tail) + red_end_of_idle_period(&q->vars); + + sch_tree_unlock(sch); + choke_free(old); + return 0; +} + +static int choke_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + return choke_change(sch, opt, extack); +} + +static int choke_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct choke_sched_data *q = qdisc_priv(sch); + struct nlattr *opts = NULL; + struct tc_red_qopt opt = { + .limit = q->limit, + .flags = q->flags, + .qth_min = q->parms.qth_min >> q->parms.Wlog, + .qth_max = q->parms.qth_max >> q->parms.Wlog, + .Wlog = q->parms.Wlog, + .Plog = q->parms.Plog, + .Scell_log = q->parms.Scell_log, + }; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + + if (nla_put(skb, TCA_CHOKE_PARMS, sizeof(opt), &opt) || + nla_put_u32(skb, TCA_CHOKE_MAX_P, q->parms.max_P)) + goto nla_put_failure; + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -EMSGSIZE; +} + +static int choke_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct choke_sched_data *q = qdisc_priv(sch); + struct tc_choke_xstats st = { + .early = q->stats.prob_drop + q->stats.forced_drop, + .marked = q->stats.prob_mark + q->stats.forced_mark, + .pdrop = q->stats.pdrop, + .other = q->stats.other, + .matched = q->stats.matched, + }; + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static void choke_destroy(struct Qdisc *sch) +{ + struct choke_sched_data *q = qdisc_priv(sch); + + choke_free(q->tab); +} + +static struct sk_buff *choke_peek_head(struct Qdisc *sch) +{ + struct choke_sched_data *q = qdisc_priv(sch); + + return (q->head != q->tail) ? q->tab[q->head] : NULL; +} + +static struct Qdisc_ops choke_qdisc_ops __read_mostly = { + .id = "choke", + .priv_size = sizeof(struct choke_sched_data), + + .enqueue = choke_enqueue, + .dequeue = choke_dequeue, + .peek = choke_peek_head, + .init = choke_init, + .destroy = choke_destroy, + .reset = choke_reset, + .change = choke_change, + .dump = choke_dump, + .dump_stats = choke_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init choke_module_init(void) +{ + return register_qdisc(&choke_qdisc_ops); +} + +static void __exit choke_module_exit(void) +{ + unregister_qdisc(&choke_qdisc_ops); +} + +module_init(choke_module_init) +module_exit(choke_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c new file mode 100644 index 000000000..77fae0b7c --- /dev/null +++ b/net/sched/sch_codel.c @@ -0,0 +1,309 @@ +/* + * Codel - The Controlled-Delay Active Queue Management algorithm + * + * Copyright (C) 2011-2012 Kathleen Nichols <nichols@pollere.com> + * Copyright (C) 2011-2012 Van Jacobson <van@pollere.net> + * + * Implemented on linux by : + * Copyright (C) 2012 Michael D. Taht <dave.taht@bufferbloat.net> + * Copyright (C) 2012,2015 Eric Dumazet <edumazet@google.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions, and the following disclaimer, + * without modification. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. The names of the authors may not be used to endorse or promote products + * derived from this software without specific prior written permission. + * + * Alternatively, provided that this notice is retained in full, this + * software may be distributed under the terms of the GNU General + * Public License ("GPL") version 2, in which case the provisions of the + * GPL apply INSTEAD OF those given above. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH + * DAMAGE. + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/prefetch.h> +#include <net/pkt_sched.h> +#include <net/codel.h> +#include <net/codel_impl.h> +#include <net/codel_qdisc.h> + + +#define DEFAULT_CODEL_LIMIT 1000 + +struct codel_sched_data { + struct codel_params params; + struct codel_vars vars; + struct codel_stats stats; + u32 drop_overlimit; +}; + +/* This is the specific function called from codel_dequeue() + * to dequeue a packet from queue. Note: backlog is handled in + * codel, we dont need to reduce it here. + */ +static struct sk_buff *dequeue_func(struct codel_vars *vars, void *ctx) +{ + struct Qdisc *sch = ctx; + struct sk_buff *skb = __qdisc_dequeue_head(&sch->q); + + if (skb) { + sch->qstats.backlog -= qdisc_pkt_len(skb); + prefetch(&skb->end); /* we'll need skb_shinfo() */ + } + return skb; +} + +static void drop_func(struct sk_buff *skb, void *ctx) +{ + struct Qdisc *sch = ctx; + + kfree_skb(skb); + qdisc_qstats_drop(sch); +} + +static struct sk_buff *codel_qdisc_dequeue(struct Qdisc *sch) +{ + struct codel_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + + skb = codel_dequeue(sch, &sch->qstats.backlog, &q->params, &q->vars, + &q->stats, qdisc_pkt_len, codel_get_enqueue_time, + drop_func, dequeue_func); + + /* We cant call qdisc_tree_reduce_backlog() if our qlen is 0, + * or HTB crashes. Defer it for next round. + */ + if (q->stats.drop_count && sch->q.qlen) { + qdisc_tree_reduce_backlog(sch, q->stats.drop_count, q->stats.drop_len); + q->stats.drop_count = 0; + q->stats.drop_len = 0; + } + if (skb) + qdisc_bstats_update(sch, skb); + return skb; +} + +static int codel_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct codel_sched_data *q; + + if (likely(qdisc_qlen(sch) < sch->limit)) { + codel_set_enqueue_time(skb); + return qdisc_enqueue_tail(skb, sch); + } + q = qdisc_priv(sch); + q->drop_overlimit++; + return qdisc_drop(skb, sch, to_free); +} + +static const struct nla_policy codel_policy[TCA_CODEL_MAX + 1] = { + [TCA_CODEL_TARGET] = { .type = NLA_U32 }, + [TCA_CODEL_LIMIT] = { .type = NLA_U32 }, + [TCA_CODEL_INTERVAL] = { .type = NLA_U32 }, + [TCA_CODEL_ECN] = { .type = NLA_U32 }, + [TCA_CODEL_CE_THRESHOLD]= { .type = NLA_U32 }, +}; + +static int codel_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct codel_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_CODEL_MAX + 1]; + unsigned int qlen, dropped = 0; + int err; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_CODEL_MAX, opt, codel_policy, NULL); + if (err < 0) + return err; + + sch_tree_lock(sch); + + if (tb[TCA_CODEL_TARGET]) { + u32 target = nla_get_u32(tb[TCA_CODEL_TARGET]); + + q->params.target = ((u64)target * NSEC_PER_USEC) >> CODEL_SHIFT; + } + + if (tb[TCA_CODEL_CE_THRESHOLD]) { + u64 val = nla_get_u32(tb[TCA_CODEL_CE_THRESHOLD]); + + q->params.ce_threshold = (val * NSEC_PER_USEC) >> CODEL_SHIFT; + } + + if (tb[TCA_CODEL_INTERVAL]) { + u32 interval = nla_get_u32(tb[TCA_CODEL_INTERVAL]); + + q->params.interval = ((u64)interval * NSEC_PER_USEC) >> CODEL_SHIFT; + } + + if (tb[TCA_CODEL_LIMIT]) + sch->limit = nla_get_u32(tb[TCA_CODEL_LIMIT]); + + if (tb[TCA_CODEL_ECN]) + q->params.ecn = !!nla_get_u32(tb[TCA_CODEL_ECN]); + + qlen = sch->q.qlen; + while (sch->q.qlen > sch->limit) { + struct sk_buff *skb = __qdisc_dequeue_head(&sch->q); + + dropped += qdisc_pkt_len(skb); + qdisc_qstats_backlog_dec(sch, skb); + rtnl_qdisc_drop(skb, sch); + } + qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped); + + sch_tree_unlock(sch); + return 0; +} + +static int codel_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct codel_sched_data *q = qdisc_priv(sch); + + sch->limit = DEFAULT_CODEL_LIMIT; + + codel_params_init(&q->params); + codel_vars_init(&q->vars); + codel_stats_init(&q->stats); + q->params.mtu = psched_mtu(qdisc_dev(sch)); + + if (opt) { + int err = codel_change(sch, opt, extack); + + if (err) + return err; + } + + if (sch->limit >= 1) + sch->flags |= TCQ_F_CAN_BYPASS; + else + sch->flags &= ~TCQ_F_CAN_BYPASS; + + return 0; +} + +static int codel_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct codel_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_CODEL_TARGET, + codel_time_to_us(q->params.target)) || + nla_put_u32(skb, TCA_CODEL_LIMIT, + sch->limit) || + nla_put_u32(skb, TCA_CODEL_INTERVAL, + codel_time_to_us(q->params.interval)) || + nla_put_u32(skb, TCA_CODEL_ECN, + q->params.ecn)) + goto nla_put_failure; + if (q->params.ce_threshold != CODEL_DISABLED_THRESHOLD && + nla_put_u32(skb, TCA_CODEL_CE_THRESHOLD, + codel_time_to_us(q->params.ce_threshold))) + goto nla_put_failure; + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -1; +} + +static int codel_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + const struct codel_sched_data *q = qdisc_priv(sch); + struct tc_codel_xstats st = { + .maxpacket = q->stats.maxpacket, + .count = q->vars.count, + .lastcount = q->vars.lastcount, + .drop_overlimit = q->drop_overlimit, + .ldelay = codel_time_to_us(q->vars.ldelay), + .dropping = q->vars.dropping, + .ecn_mark = q->stats.ecn_mark, + .ce_mark = q->stats.ce_mark, + }; + + if (q->vars.dropping) { + codel_tdiff_t delta = q->vars.drop_next - codel_get_time(); + + if (delta >= 0) + st.drop_next = codel_time_to_us(delta); + else + st.drop_next = -codel_time_to_us(-delta); + } + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static void codel_reset(struct Qdisc *sch) +{ + struct codel_sched_data *q = qdisc_priv(sch); + + qdisc_reset_queue(sch); + codel_vars_init(&q->vars); +} + +static struct Qdisc_ops codel_qdisc_ops __read_mostly = { + .id = "codel", + .priv_size = sizeof(struct codel_sched_data), + + .enqueue = codel_qdisc_enqueue, + .dequeue = codel_qdisc_dequeue, + .peek = qdisc_peek_dequeued, + .init = codel_init, + .reset = codel_reset, + .change = codel_change, + .dump = codel_dump, + .dump_stats = codel_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init codel_module_init(void) +{ + return register_qdisc(&codel_qdisc_ops); +} + +static void __exit codel_module_exit(void) +{ + unregister_qdisc(&codel_qdisc_ops); +} + +module_init(codel_module_init) +module_exit(codel_module_exit) + +MODULE_DESCRIPTION("Controlled Delay queue discipline"); +MODULE_AUTHOR("Dave Taht"); +MODULE_AUTHOR("Eric Dumazet"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c new file mode 100644 index 000000000..cdebaed0f --- /dev/null +++ b/net/sched/sch_drr.c @@ -0,0 +1,514 @@ +/* + * net/sched/sch_drr.c Deficit Round Robin scheduler + * + * Copyright (c) 2008 Patrick McHardy <kaber@trash.net> + * + * 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/slab.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/netdevice.h> +#include <linux/pkt_sched.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +struct drr_class { + struct Qdisc_class_common common; + unsigned int filter_cnt; + + struct gnet_stats_basic_packed bstats; + struct gnet_stats_queue qstats; + struct net_rate_estimator __rcu *rate_est; + struct list_head alist; + struct Qdisc *qdisc; + + u32 quantum; + u32 deficit; +}; + +struct drr_sched { + struct list_head active; + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + struct Qdisc_class_hash clhash; +}; + +static struct drr_class *drr_find_class(struct Qdisc *sch, u32 classid) +{ + struct drr_sched *q = qdisc_priv(sch); + struct Qdisc_class_common *clc; + + clc = qdisc_class_find(&q->clhash, classid); + if (clc == NULL) + return NULL; + return container_of(clc, struct drr_class, common); +} + +static void drr_purge_queue(struct drr_class *cl) +{ + unsigned int len = cl->qdisc->q.qlen; + unsigned int backlog = cl->qdisc->qstats.backlog; + + qdisc_reset(cl->qdisc); + qdisc_tree_reduce_backlog(cl->qdisc, len, backlog); +} + +static const struct nla_policy drr_policy[TCA_DRR_MAX + 1] = { + [TCA_DRR_QUANTUM] = { .type = NLA_U32 }, +}; + +static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid, + struct nlattr **tca, unsigned long *arg, + struct netlink_ext_ack *extack) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl = (struct drr_class *)*arg; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_DRR_MAX + 1]; + u32 quantum; + int err; + + if (!opt) { + NL_SET_ERR_MSG(extack, "DRR options are required for this operation"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_DRR_MAX, opt, drr_policy, extack); + if (err < 0) + return err; + + if (tb[TCA_DRR_QUANTUM]) { + quantum = nla_get_u32(tb[TCA_DRR_QUANTUM]); + if (quantum == 0) { + NL_SET_ERR_MSG(extack, "Specified DRR quantum cannot be zero"); + return -EINVAL; + } + } else + quantum = psched_mtu(qdisc_dev(sch)); + + if (cl != NULL) { + if (tca[TCA_RATE]) { + err = gen_replace_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) { + NL_SET_ERR_MSG(extack, "Failed to replace estimator"); + return err; + } + } + + sch_tree_lock(sch); + if (tb[TCA_DRR_QUANTUM]) + cl->quantum = quantum; + sch_tree_unlock(sch); + + return 0; + } + + cl = kzalloc(sizeof(struct drr_class), GFP_KERNEL); + if (cl == NULL) + return -ENOBUFS; + + cl->common.classid = classid; + cl->quantum = quantum; + cl->qdisc = qdisc_create_dflt(sch->dev_queue, + &pfifo_qdisc_ops, classid, + NULL); + if (cl->qdisc == NULL) + cl->qdisc = &noop_qdisc; + else + qdisc_hash_add(cl->qdisc, true); + + if (tca[TCA_RATE]) { + err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) { + NL_SET_ERR_MSG(extack, "Failed to replace estimator"); + qdisc_put(cl->qdisc); + kfree(cl); + return err; + } + } + + sch_tree_lock(sch); + qdisc_class_hash_insert(&q->clhash, &cl->common); + sch_tree_unlock(sch); + + qdisc_class_hash_grow(sch, &q->clhash); + + *arg = (unsigned long)cl; + return 0; +} + +static void drr_destroy_class(struct Qdisc *sch, struct drr_class *cl) +{ + gen_kill_estimator(&cl->rate_est); + qdisc_put(cl->qdisc); + kfree(cl); +} + +static int drr_delete_class(struct Qdisc *sch, unsigned long arg) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl = (struct drr_class *)arg; + + if (cl->filter_cnt > 0) + return -EBUSY; + + sch_tree_lock(sch); + + drr_purge_queue(cl); + qdisc_class_hash_remove(&q->clhash, &cl->common); + + sch_tree_unlock(sch); + + drr_destroy_class(sch, cl); + return 0; +} + +static unsigned long drr_search_class(struct Qdisc *sch, u32 classid) +{ + return (unsigned long)drr_find_class(sch, classid); +} + +static struct tcf_block *drr_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct drr_sched *q = qdisc_priv(sch); + + if (cl) { + NL_SET_ERR_MSG(extack, "DRR classid must be zero"); + return NULL; + } + + return q->block; +} + +static unsigned long drr_bind_tcf(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + struct drr_class *cl = drr_find_class(sch, classid); + + if (cl != NULL) + cl->filter_cnt++; + + return (unsigned long)cl; +} + +static void drr_unbind_tcf(struct Qdisc *sch, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + cl->filter_cnt--; +} + +static int drr_graft_class(struct Qdisc *sch, unsigned long arg, + struct Qdisc *new, struct Qdisc **old, + struct netlink_ext_ack *extack) +{ + struct drr_class *cl = (struct drr_class *)arg; + + if (new == NULL) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + cl->common.classid, NULL); + if (new == NULL) + new = &noop_qdisc; + } + + *old = qdisc_replace(sch, new, &cl->qdisc); + return 0; +} + +static struct Qdisc *drr_class_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + return cl->qdisc; +} + +static void drr_qlen_notify(struct Qdisc *csh, unsigned long arg) +{ + struct drr_class *cl = (struct drr_class *)arg; + + list_del(&cl->alist); +} + +static int drr_dump_class(struct Qdisc *sch, unsigned long arg, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct drr_class *cl = (struct drr_class *)arg; + struct nlattr *nest; + + tcm->tcm_parent = TC_H_ROOT; + tcm->tcm_handle = cl->common.classid; + tcm->tcm_info = cl->qdisc->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + if (nla_put_u32(skb, TCA_DRR_QUANTUM, cl->quantum)) + goto nla_put_failure; + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg, + struct gnet_dump *d) +{ + struct drr_class *cl = (struct drr_class *)arg; + __u32 qlen = cl->qdisc->q.qlen; + struct tc_drr_stats xstats; + + memset(&xstats, 0, sizeof(xstats)); + if (qlen) + xstats.deficit = cl->deficit; + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || + gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || + gnet_stats_copy_queue(d, NULL, &cl->qdisc->qstats, qlen) < 0) + return -1; + + return gnet_stats_copy_app(d, &xstats, sizeof(xstats)); +} + +static void drr_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, (unsigned long)cl, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static struct drr_class *drr_classify(struct sk_buff *skb, struct Qdisc *sch, + int *qerr) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct tcf_result res; + struct tcf_proto *fl; + int result; + + if (TC_H_MAJ(skb->priority ^ sch->handle) == 0) { + cl = drr_find_class(sch, skb->priority); + if (cl != NULL) + return cl; + } + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + fl = rcu_dereference_bh(q->filter_list); + result = tcf_classify(skb, fl, &res, false); + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + } +#endif + cl = (struct drr_class *)res.class; + if (cl == NULL) + cl = drr_find_class(sch, res.classid); + return cl; + } + return NULL; +} + +static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + int err = 0; + + cl = drr_classify(skb, sch, &err); + if (cl == NULL) { + if (err & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return err; + } + + err = qdisc_enqueue(skb, cl->qdisc, to_free); + if (unlikely(err != NET_XMIT_SUCCESS)) { + if (net_xmit_drop_count(err)) { + cl->qstats.drops++; + qdisc_qstats_drop(sch); + } + return err; + } + + if (cl->qdisc->q.qlen == 1) { + list_add_tail(&cl->alist, &q->active); + cl->deficit = cl->quantum; + } + + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + return err; +} + +static struct sk_buff *drr_dequeue(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct sk_buff *skb; + unsigned int len; + + if (list_empty(&q->active)) + goto out; + while (1) { + cl = list_first_entry(&q->active, struct drr_class, alist); + skb = cl->qdisc->ops->peek(cl->qdisc); + if (skb == NULL) { + qdisc_warn_nonwc(__func__, cl->qdisc); + goto out; + } + + len = qdisc_pkt_len(skb); + if (len <= cl->deficit) { + cl->deficit -= len; + skb = qdisc_dequeue_peeked(cl->qdisc); + if (unlikely(skb == NULL)) + goto out; + if (cl->qdisc->q.qlen == 0) + list_del(&cl->alist); + + bstats_update(&cl->bstats, skb); + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + return skb; + } + + cl->deficit += cl->quantum; + list_move_tail(&cl->alist, &q->active); + } +out: + return NULL; +} + +static int drr_init_qdisc(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct drr_sched *q = qdisc_priv(sch); + int err; + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + err = qdisc_class_hash_init(&q->clhash); + if (err < 0) + return err; + INIT_LIST_HEAD(&q->active); + return 0; +} + +static void drr_reset_qdisc(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + unsigned int i; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + if (cl->qdisc->q.qlen) + list_del(&cl->alist); + qdisc_reset(cl->qdisc); + } + } + sch->qstats.backlog = 0; + sch->q.qlen = 0; +} + +static void drr_destroy_qdisc(struct Qdisc *sch) +{ + struct drr_sched *q = qdisc_priv(sch); + struct drr_class *cl; + struct hlist_node *next; + unsigned int i; + + tcf_block_put(q->block); + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], + common.hnode) + drr_destroy_class(sch, cl); + } + qdisc_class_hash_destroy(&q->clhash); +} + +static const struct Qdisc_class_ops drr_class_ops = { + .change = drr_change_class, + .delete = drr_delete_class, + .find = drr_search_class, + .tcf_block = drr_tcf_block, + .bind_tcf = drr_bind_tcf, + .unbind_tcf = drr_unbind_tcf, + .graft = drr_graft_class, + .leaf = drr_class_leaf, + .qlen_notify = drr_qlen_notify, + .dump = drr_dump_class, + .dump_stats = drr_dump_class_stats, + .walk = drr_walk, +}; + +static struct Qdisc_ops drr_qdisc_ops __read_mostly = { + .cl_ops = &drr_class_ops, + .id = "drr", + .priv_size = sizeof(struct drr_sched), + .enqueue = drr_enqueue, + .dequeue = drr_dequeue, + .peek = qdisc_peek_dequeued, + .init = drr_init_qdisc, + .reset = drr_reset_qdisc, + .destroy = drr_destroy_qdisc, + .owner = THIS_MODULE, +}; + +static int __init drr_init(void) +{ + return register_qdisc(&drr_qdisc_ops); +} + +static void __exit drr_exit(void) +{ + unregister_qdisc(&drr_qdisc_ops); +} + +module_init(drr_init); +module_exit(drr_exit); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c new file mode 100644 index 000000000..47a61689d --- /dev/null +++ b/net/sched/sch_dsmark.c @@ -0,0 +1,519 @@ +/* net/sched/sch_dsmark.c - Differentiated Services field marker */ + +/* Written 1998-2000 by Werner Almesberger, EPFL ICA */ + + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/bitops.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/dsfield.h> +#include <net/inet_ecn.h> +#include <asm/byteorder.h> + +/* + * classid class marking + * ------- ----- ------- + * n/a 0 n/a + * x:0 1 use entry [0] + * ... ... ... + * x:y y>0 y+1 use entry [y] + * ... ... ... + * x:indices-1 indices use entry [indices-1] + * ... ... ... + * x:y y+1 use entry [y & (indices-1)] + * ... ... ... + * 0xffff 0x10000 use entry [indices-1] + */ + + +#define NO_DEFAULT_INDEX (1 << 16) + +struct mask_value { + u8 mask; + u8 value; +}; + +struct dsmark_qdisc_data { + struct Qdisc *q; + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + struct mask_value *mv; + u16 indices; + u8 set_tc_index; + u32 default_index; /* index range is 0...0xffff */ +#define DSMARK_EMBEDDED_SZ 16 + struct mask_value embedded[DSMARK_EMBEDDED_SZ]; +}; + +static inline int dsmark_valid_index(struct dsmark_qdisc_data *p, u16 index) +{ + return index <= p->indices && index > 0; +} + +/* ------------------------- Class/flow operations ------------------------- */ + +static int dsmark_graft(struct Qdisc *sch, unsigned long arg, + struct Qdisc *new, struct Qdisc **old, + struct netlink_ext_ack *extack) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + + pr_debug("%s(sch %p,[qdisc %p],new %p,old %p)\n", + __func__, sch, p, new, old); + + if (new == NULL) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + sch->handle, NULL); + if (new == NULL) + new = &noop_qdisc; + } + + *old = qdisc_replace(sch, new, &p->q); + return 0; +} + +static struct Qdisc *dsmark_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + return p->q; +} + +static unsigned long dsmark_find(struct Qdisc *sch, u32 classid) +{ + return TC_H_MIN(classid) + 1; +} + +static unsigned long dsmark_bind_filter(struct Qdisc *sch, + unsigned long parent, u32 classid) +{ + pr_debug("%s(sch %p,[qdisc %p],classid %x)\n", + __func__, sch, qdisc_priv(sch), classid); + + return dsmark_find(sch, classid); +} + +static void dsmark_unbind_filter(struct Qdisc *sch, unsigned long cl) +{ +} + +static const struct nla_policy dsmark_policy[TCA_DSMARK_MAX + 1] = { + [TCA_DSMARK_INDICES] = { .type = NLA_U16 }, + [TCA_DSMARK_DEFAULT_INDEX] = { .type = NLA_U16 }, + [TCA_DSMARK_SET_TC_INDEX] = { .type = NLA_FLAG }, + [TCA_DSMARK_MASK] = { .type = NLA_U8 }, + [TCA_DSMARK_VALUE] = { .type = NLA_U8 }, +}; + +static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent, + struct nlattr **tca, unsigned long *arg, + struct netlink_ext_ack *extack) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_DSMARK_MAX + 1]; + int err = -EINVAL; + + pr_debug("%s(sch %p,[qdisc %p],classid %x,parent %x), arg 0x%lx\n", + __func__, sch, p, classid, parent, *arg); + + if (!dsmark_valid_index(p, *arg)) { + err = -ENOENT; + goto errout; + } + + if (!opt) + goto errout; + + err = nla_parse_nested(tb, TCA_DSMARK_MAX, opt, dsmark_policy, NULL); + if (err < 0) + goto errout; + + if (tb[TCA_DSMARK_VALUE]) + p->mv[*arg - 1].value = nla_get_u8(tb[TCA_DSMARK_VALUE]); + + if (tb[TCA_DSMARK_MASK]) + p->mv[*arg - 1].mask = nla_get_u8(tb[TCA_DSMARK_MASK]); + + err = 0; + +errout: + return err; +} + +static int dsmark_delete(struct Qdisc *sch, unsigned long arg) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + + if (!dsmark_valid_index(p, arg)) + return -EINVAL; + + p->mv[arg - 1].mask = 0xff; + p->mv[arg - 1].value = 0; + + return 0; +} + +static void dsmark_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + int i; + + pr_debug("%s(sch %p,[qdisc %p],walker %p)\n", + __func__, sch, p, walker); + + if (walker->stop) + return; + + for (i = 0; i < p->indices; i++) { + if (p->mv[i].mask == 0xff && !p->mv[i].value) + goto ignore; + if (walker->count >= walker->skip) { + if (walker->fn(sch, i + 1, walker) < 0) { + walker->stop = 1; + break; + } + } +ignore: + walker->count++; + } +} + +static struct tcf_block *dsmark_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + + return p->block; +} + +/* --------------------------- Qdisc operations ---------------------------- */ + +static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + int err; + + pr_debug("%s(skb %p,sch %p,[qdisc %p])\n", __func__, skb, sch, p); + + if (p->set_tc_index) { + int wlen = skb_network_offset(skb); + + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + wlen += sizeof(struct iphdr); + if (!pskb_may_pull(skb, wlen) || + skb_try_make_writable(skb, wlen)) + goto drop; + + skb->tc_index = ipv4_get_dsfield(ip_hdr(skb)) + & ~INET_ECN_MASK; + break; + + case htons(ETH_P_IPV6): + wlen += sizeof(struct ipv6hdr); + if (!pskb_may_pull(skb, wlen) || + skb_try_make_writable(skb, wlen)) + goto drop; + + skb->tc_index = ipv6_get_dsfield(ipv6_hdr(skb)) + & ~INET_ECN_MASK; + break; + default: + skb->tc_index = 0; + break; + } + } + + if (TC_H_MAJ(skb->priority) == sch->handle) + skb->tc_index = TC_H_MIN(skb->priority); + else { + struct tcf_result res; + struct tcf_proto *fl = rcu_dereference_bh(p->filter_list); + int result = tcf_classify(skb, fl, &res, false); + + pr_debug("result %d class 0x%04x\n", result, res.classid); + + switch (result) { +#ifdef CONFIG_NET_CLS_ACT + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + __qdisc_drop(skb, to_free); + return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + + case TC_ACT_SHOT: + goto drop; +#endif + case TC_ACT_OK: + skb->tc_index = TC_H_MIN(res.classid); + break; + + default: + if (p->default_index != NO_DEFAULT_INDEX) + skb->tc_index = p->default_index; + break; + } + } + + err = qdisc_enqueue(skb, p->q, to_free); + if (err != NET_XMIT_SUCCESS) { + if (net_xmit_drop_count(err)) + qdisc_qstats_drop(sch); + return err; + } + + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + + return NET_XMIT_SUCCESS; + +drop: + qdisc_drop(skb, sch, to_free); + return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; +} + +static struct sk_buff *dsmark_dequeue(struct Qdisc *sch) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + struct sk_buff *skb; + u32 index; + + pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); + + skb = qdisc_dequeue_peeked(p->q); + if (skb == NULL) + return NULL; + + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + + index = skb->tc_index & (p->indices - 1); + pr_debug("index %d->%d\n", skb->tc_index, index); + + switch (skb_protocol(skb, true)) { + case htons(ETH_P_IP): + ipv4_change_dsfield(ip_hdr(skb), p->mv[index].mask, + p->mv[index].value); + break; + case htons(ETH_P_IPV6): + ipv6_change_dsfield(ipv6_hdr(skb), p->mv[index].mask, + p->mv[index].value); + break; + default: + /* + * Only complain if a change was actually attempted. + * This way, we can send non-IP traffic through dsmark + * and don't need yet another qdisc as a bypass. + */ + if (p->mv[index].mask != 0xff || p->mv[index].value) + pr_warn("%s: unsupported protocol %d\n", + __func__, ntohs(skb_protocol(skb, true))); + break; + } + + return skb; +} + +static struct sk_buff *dsmark_peek(struct Qdisc *sch) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + + pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); + + return p->q->ops->peek(p->q); +} + +static int dsmark_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + struct nlattr *tb[TCA_DSMARK_MAX + 1]; + int err = -EINVAL; + u32 default_index = NO_DEFAULT_INDEX; + u16 indices; + int i; + + pr_debug("%s(sch %p,[qdisc %p],opt %p)\n", __func__, sch, p, opt); + + if (!opt) + goto errout; + + err = tcf_block_get(&p->block, &p->filter_list, sch, extack); + if (err) + return err; + + err = nla_parse_nested(tb, TCA_DSMARK_MAX, opt, dsmark_policy, NULL); + if (err < 0) + goto errout; + + err = -EINVAL; + if (!tb[TCA_DSMARK_INDICES]) + goto errout; + indices = nla_get_u16(tb[TCA_DSMARK_INDICES]); + + if (hweight32(indices) != 1) + goto errout; + + if (tb[TCA_DSMARK_DEFAULT_INDEX]) + default_index = nla_get_u16(tb[TCA_DSMARK_DEFAULT_INDEX]); + + if (indices <= DSMARK_EMBEDDED_SZ) + p->mv = p->embedded; + else + p->mv = kmalloc_array(indices, sizeof(*p->mv), GFP_KERNEL); + if (!p->mv) { + err = -ENOMEM; + goto errout; + } + for (i = 0; i < indices; i++) { + p->mv[i].mask = 0xff; + p->mv[i].value = 0; + } + p->indices = indices; + p->default_index = default_index; + p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]); + + p->q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, sch->handle, + NULL); + if (p->q == NULL) + p->q = &noop_qdisc; + else + qdisc_hash_add(p->q, true); + + pr_debug("%s: qdisc %p\n", __func__, p->q); + + err = 0; +errout: + return err; +} + +static void dsmark_reset(struct Qdisc *sch) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + + pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); + if (p->q) + qdisc_reset(p->q); + sch->qstats.backlog = 0; + sch->q.qlen = 0; +} + +static void dsmark_destroy(struct Qdisc *sch) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + + pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); + + tcf_block_put(p->block); + qdisc_put(p->q); + if (p->mv != p->embedded) + kfree(p->mv); +} + +static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + struct nlattr *opts = NULL; + + pr_debug("%s(sch %p,[qdisc %p],class %ld\n", __func__, sch, p, cl); + + if (!dsmark_valid_index(p, cl)) + return -EINVAL; + + tcm->tcm_handle = TC_H_MAKE(TC_H_MAJ(sch->handle), cl - 1); + tcm->tcm_info = p->q->handle; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + if (nla_put_u8(skb, TCA_DSMARK_MASK, p->mv[cl - 1].mask) || + nla_put_u8(skb, TCA_DSMARK_VALUE, p->mv[cl - 1].value)) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -EMSGSIZE; +} + +static int dsmark_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct dsmark_qdisc_data *p = qdisc_priv(sch); + struct nlattr *opts = NULL; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + if (nla_put_u16(skb, TCA_DSMARK_INDICES, p->indices)) + goto nla_put_failure; + + if (p->default_index != NO_DEFAULT_INDEX && + nla_put_u16(skb, TCA_DSMARK_DEFAULT_INDEX, p->default_index)) + goto nla_put_failure; + + if (p->set_tc_index && + nla_put_flag(skb, TCA_DSMARK_SET_TC_INDEX)) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -EMSGSIZE; +} + +static const struct Qdisc_class_ops dsmark_class_ops = { + .graft = dsmark_graft, + .leaf = dsmark_leaf, + .find = dsmark_find, + .change = dsmark_change, + .delete = dsmark_delete, + .walk = dsmark_walk, + .tcf_block = dsmark_tcf_block, + .bind_tcf = dsmark_bind_filter, + .unbind_tcf = dsmark_unbind_filter, + .dump = dsmark_dump_class, +}; + +static struct Qdisc_ops dsmark_qdisc_ops __read_mostly = { + .next = NULL, + .cl_ops = &dsmark_class_ops, + .id = "dsmark", + .priv_size = sizeof(struct dsmark_qdisc_data), + .enqueue = dsmark_enqueue, + .dequeue = dsmark_dequeue, + .peek = dsmark_peek, + .init = dsmark_init, + .reset = dsmark_reset, + .destroy = dsmark_destroy, + .change = NULL, + .dump = dsmark_dump, + .owner = THIS_MODULE, +}; + +static int __init dsmark_module_init(void) +{ + return register_qdisc(&dsmark_qdisc_ops); +} + +static void __exit dsmark_module_exit(void) +{ + unregister_qdisc(&dsmark_qdisc_ops); +} + +module_init(dsmark_module_init) +module_exit(dsmark_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_etf.c b/net/sched/sch_etf.c new file mode 100644 index 000000000..2278f3d42 --- /dev/null +++ b/net/sched/sch_etf.c @@ -0,0 +1,485 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* net/sched/sch_etf.c Earliest TxTime First queueing discipline. + * + * Authors: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com> + * Vinicius Costa Gomes <vinicius.gomes@intel.com> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/errqueue.h> +#include <linux/rbtree.h> +#include <linux/skbuff.h> +#include <linux/posix-timers.h> +#include <net/netlink.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> +#include <net/sock.h> + +#define DEADLINE_MODE_IS_ON(x) ((x)->flags & TC_ETF_DEADLINE_MODE_ON) +#define OFFLOAD_IS_ON(x) ((x)->flags & TC_ETF_OFFLOAD_ON) + +struct etf_sched_data { + bool offload; + bool deadline_mode; + int clockid; + int queue; + s32 delta; /* in ns */ + ktime_t last; /* The txtime of the last skb sent to the netdevice. */ + struct rb_root head; + struct qdisc_watchdog watchdog; + ktime_t (*get_time)(void); +}; + +static const struct nla_policy etf_policy[TCA_ETF_MAX + 1] = { + [TCA_ETF_PARMS] = { .len = sizeof(struct tc_etf_qopt) }, +}; + +static inline int validate_input_params(struct tc_etf_qopt *qopt, + struct netlink_ext_ack *extack) +{ + /* Check if params comply to the following rules: + * * Clockid and delta must be valid. + * + * * Dynamic clockids are not supported. + * + * * Delta must be a positive integer. + * + * Also note that for the HW offload case, we must + * expect that system clocks have been synchronized to PHC. + */ + if (qopt->clockid < 0) { + NL_SET_ERR_MSG(extack, "Dynamic clockids are not supported"); + return -ENOTSUPP; + } + + if (qopt->clockid != CLOCK_TAI) { + NL_SET_ERR_MSG(extack, "Invalid clockid. CLOCK_TAI must be used"); + return -EINVAL; + } + + if (qopt->delta < 0) { + NL_SET_ERR_MSG(extack, "Delta must be positive"); + return -EINVAL; + } + + return 0; +} + +static bool is_packet_valid(struct Qdisc *sch, struct sk_buff *nskb) +{ + struct etf_sched_data *q = qdisc_priv(sch); + ktime_t txtime = nskb->tstamp; + struct sock *sk = nskb->sk; + ktime_t now; + + if (!sk || !sk_fullsock(sk)) + return false; + + if (!sock_flag(sk, SOCK_TXTIME)) + return false; + + /* We don't perform crosstimestamping. + * Drop if packet's clockid differs from qdisc's. + */ + if (sk->sk_clockid != q->clockid) + return false; + + if (sk->sk_txtime_deadline_mode != q->deadline_mode) + return false; + + now = q->get_time(); + if (ktime_before(txtime, now) || ktime_before(txtime, q->last)) + return false; + + return true; +} + +static struct sk_buff *etf_peek_timesortedlist(struct Qdisc *sch) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct rb_node *p; + + p = rb_first(&q->head); + if (!p) + return NULL; + + return rb_to_skb(p); +} + +static void reset_watchdog(struct Qdisc *sch) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb = etf_peek_timesortedlist(sch); + ktime_t next; + + if (!skb) + return; + + next = ktime_sub_ns(skb->tstamp, q->delta); + qdisc_watchdog_schedule_ns(&q->watchdog, ktime_to_ns(next)); +} + +static void report_sock_error(struct sk_buff *skb, u32 err, u8 code) +{ + struct sock_exterr_skb *serr; + struct sk_buff *clone; + ktime_t txtime = skb->tstamp; + struct sock *sk = skb->sk; + + if (!sk || !sk_fullsock(sk) || !(sk->sk_txtime_report_errors)) + return; + + clone = skb_clone(skb, GFP_ATOMIC); + if (!clone) + return; + + serr = SKB_EXT_ERR(clone); + serr->ee.ee_errno = err; + serr->ee.ee_origin = SO_EE_ORIGIN_TXTIME; + serr->ee.ee_type = 0; + serr->ee.ee_code = code; + serr->ee.ee_pad = 0; + serr->ee.ee_data = (txtime >> 32); /* high part of tstamp */ + serr->ee.ee_info = txtime; /* low part of tstamp */ + + if (sock_queue_err_skb(sk, clone)) + kfree_skb(clone); +} + +static int etf_enqueue_timesortedlist(struct sk_buff *nskb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct rb_node **p = &q->head.rb_node, *parent = NULL; + ktime_t txtime = nskb->tstamp; + + if (!is_packet_valid(sch, nskb)) { + report_sock_error(nskb, EINVAL, + SO_EE_CODE_TXTIME_INVALID_PARAM); + return qdisc_drop(nskb, sch, to_free); + } + + while (*p) { + struct sk_buff *skb; + + parent = *p; + skb = rb_to_skb(parent); + if (ktime_after(txtime, skb->tstamp)) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&nskb->rbnode, parent, p); + rb_insert_color(&nskb->rbnode, &q->head); + + qdisc_qstats_backlog_inc(sch, nskb); + sch->q.qlen++; + + /* Now we may need to re-arm the qdisc watchdog for the next packet. */ + reset_watchdog(sch); + + return NET_XMIT_SUCCESS; +} + +static void timesortedlist_erase(struct Qdisc *sch, struct sk_buff *skb, + bool drop) +{ + struct etf_sched_data *q = qdisc_priv(sch); + + rb_erase(&skb->rbnode, &q->head); + + /* The rbnode field in the skb re-uses these fields, now that + * we are done with the rbnode, reset them. + */ + skb->next = NULL; + skb->prev = NULL; + skb->dev = qdisc_dev(sch); + + qdisc_qstats_backlog_dec(sch, skb); + + if (drop) { + struct sk_buff *to_free = NULL; + + report_sock_error(skb, ECANCELED, SO_EE_CODE_TXTIME_MISSED); + + qdisc_drop(skb, sch, &to_free); + kfree_skb_list(to_free); + qdisc_qstats_overlimit(sch); + } else { + qdisc_bstats_update(sch, skb); + + q->last = skb->tstamp; + } + + sch->q.qlen--; +} + +static struct sk_buff *etf_dequeue_timesortedlist(struct Qdisc *sch) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + ktime_t now, next; + + skb = etf_peek_timesortedlist(sch); + if (!skb) + return NULL; + + now = q->get_time(); + + /* Drop if packet has expired while in queue. */ + if (ktime_before(skb->tstamp, now)) { + timesortedlist_erase(sch, skb, true); + skb = NULL; + goto out; + } + + /* When in deadline mode, dequeue as soon as possible and change the + * txtime from deadline to (now + delta). + */ + if (q->deadline_mode) { + timesortedlist_erase(sch, skb, false); + skb->tstamp = now; + goto out; + } + + next = ktime_sub_ns(skb->tstamp, q->delta); + + /* Dequeue only if now is within the [txtime - delta, txtime] range. */ + if (ktime_after(now, next)) + timesortedlist_erase(sch, skb, false); + else + skb = NULL; + +out: + /* Now we may need to re-arm the qdisc watchdog for the next packet. */ + reset_watchdog(sch); + + return skb; +} + +static void etf_disable_offload(struct net_device *dev, + struct etf_sched_data *q) +{ + struct tc_etf_qopt_offload etf = { }; + const struct net_device_ops *ops; + int err; + + if (!q->offload) + return; + + ops = dev->netdev_ops; + if (!ops->ndo_setup_tc) + return; + + etf.queue = q->queue; + etf.enable = 0; + + err = ops->ndo_setup_tc(dev, TC_SETUP_QDISC_ETF, &etf); + if (err < 0) + pr_warn("Couldn't disable ETF offload for queue %d\n", + etf.queue); +} + +static int etf_enable_offload(struct net_device *dev, struct etf_sched_data *q, + struct netlink_ext_ack *extack) +{ + const struct net_device_ops *ops = dev->netdev_ops; + struct tc_etf_qopt_offload etf = { }; + int err; + + if (q->offload) + return 0; + + if (!ops->ndo_setup_tc) { + NL_SET_ERR_MSG(extack, "Specified device does not support ETF offload"); + return -EOPNOTSUPP; + } + + etf.queue = q->queue; + etf.enable = 1; + + err = ops->ndo_setup_tc(dev, TC_SETUP_QDISC_ETF, &etf); + if (err < 0) { + NL_SET_ERR_MSG(extack, "Specified device failed to setup ETF hardware offload"); + return err; + } + + return 0; +} + +static int etf_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + struct nlattr *tb[TCA_ETF_MAX + 1]; + struct tc_etf_qopt *qopt; + int err; + + if (!opt) { + NL_SET_ERR_MSG(extack, + "Missing ETF qdisc options which are mandatory"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_ETF_MAX, opt, etf_policy, extack); + if (err < 0) + return err; + + if (!tb[TCA_ETF_PARMS]) { + NL_SET_ERR_MSG(extack, "Missing mandatory ETF parameters"); + return -EINVAL; + } + + qopt = nla_data(tb[TCA_ETF_PARMS]); + + pr_debug("delta %d clockid %d offload %s deadline %s\n", + qopt->delta, qopt->clockid, + OFFLOAD_IS_ON(qopt) ? "on" : "off", + DEADLINE_MODE_IS_ON(qopt) ? "on" : "off"); + + err = validate_input_params(qopt, extack); + if (err < 0) + return err; + + q->queue = sch->dev_queue - netdev_get_tx_queue(dev, 0); + + if (OFFLOAD_IS_ON(qopt)) { + err = etf_enable_offload(dev, q, extack); + if (err < 0) + return err; + } + + /* Everything went OK, save the parameters used. */ + q->delta = qopt->delta; + q->clockid = qopt->clockid; + q->offload = OFFLOAD_IS_ON(qopt); + q->deadline_mode = DEADLINE_MODE_IS_ON(qopt); + + switch (q->clockid) { + case CLOCK_REALTIME: + q->get_time = ktime_get_real; + break; + case CLOCK_MONOTONIC: + q->get_time = ktime_get; + break; + case CLOCK_BOOTTIME: + q->get_time = ktime_get_boottime; + break; + case CLOCK_TAI: + q->get_time = ktime_get_clocktai; + break; + default: + NL_SET_ERR_MSG(extack, "Clockid is not supported"); + return -ENOTSUPP; + } + + qdisc_watchdog_init_clockid(&q->watchdog, sch, q->clockid); + + return 0; +} + +static void timesortedlist_clear(struct Qdisc *sch) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct rb_node *p = rb_first(&q->head); + + while (p) { + struct sk_buff *skb = rb_to_skb(p); + + p = rb_next(p); + + rb_erase(&skb->rbnode, &q->head); + rtnl_kfree_skbs(skb, skb); + sch->q.qlen--; + } +} + +static void etf_reset(struct Qdisc *sch) +{ + struct etf_sched_data *q = qdisc_priv(sch); + + /* Only cancel watchdog if it's been initialized. */ + if (q->watchdog.qdisc == sch) + qdisc_watchdog_cancel(&q->watchdog); + + /* No matter which mode we are on, it's safe to clear both lists. */ + timesortedlist_clear(sch); + __qdisc_reset_queue(&sch->q); + + sch->qstats.backlog = 0; + sch->q.qlen = 0; + + q->last = 0; +} + +static void etf_destroy(struct Qdisc *sch) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + + /* Only cancel watchdog if it's been initialized. */ + if (q->watchdog.qdisc == sch) + qdisc_watchdog_cancel(&q->watchdog); + + etf_disable_offload(dev, q); +} + +static int etf_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct etf_sched_data *q = qdisc_priv(sch); + struct tc_etf_qopt opt = { }; + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + opt.delta = q->delta; + opt.clockid = q->clockid; + if (q->offload) + opt.flags |= TC_ETF_OFFLOAD_ON; + + if (q->deadline_mode) + opt.flags |= TC_ETF_DEADLINE_MODE_ON; + + if (nla_put(skb, TCA_ETF_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static struct Qdisc_ops etf_qdisc_ops __read_mostly = { + .id = "etf", + .priv_size = sizeof(struct etf_sched_data), + .enqueue = etf_enqueue_timesortedlist, + .dequeue = etf_dequeue_timesortedlist, + .peek = etf_peek_timesortedlist, + .init = etf_init, + .reset = etf_reset, + .destroy = etf_destroy, + .dump = etf_dump, + .owner = THIS_MODULE, +}; + +static int __init etf_module_init(void) +{ + return register_qdisc(&etf_qdisc_ops); +} + +static void __exit etf_module_exit(void) +{ + unregister_qdisc(&etf_qdisc_ops); +} +module_init(etf_module_init) +module_exit(etf_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c new file mode 100644 index 000000000..3697cd799 --- /dev/null +++ b/net/sched/sch_fifo.c @@ -0,0 +1,190 @@ +/* + * net/sched/sch_fifo.c The simplest FIFO queue. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> + +/* 1 band FIFO pseudo-"scheduler" */ + +static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= sch->limit)) + return qdisc_enqueue_tail(skb, sch); + + return qdisc_drop(skb, sch, to_free); +} + +static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + if (likely(sch->q.qlen < sch->limit)) + return qdisc_enqueue_tail(skb, sch); + + return qdisc_drop(skb, sch, to_free); +} + +static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + unsigned int prev_backlog; + + if (likely(sch->q.qlen < sch->limit)) + return qdisc_enqueue_tail(skb, sch); + + prev_backlog = sch->qstats.backlog; + /* queue full, remove one skb to fulfill the limit */ + __qdisc_queue_drop_head(sch, &sch->q, to_free); + qdisc_qstats_drop(sch); + qdisc_enqueue_tail(skb, sch); + + qdisc_tree_reduce_backlog(sch, 0, prev_backlog - sch->qstats.backlog); + return NET_XMIT_CN; +} + +static int fifo_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + bool bypass; + bool is_bfifo = sch->ops == &bfifo_qdisc_ops; + + if (opt == NULL) { + u32 limit = qdisc_dev(sch)->tx_queue_len; + + if (is_bfifo) + limit *= psched_mtu(qdisc_dev(sch)); + + sch->limit = limit; + } else { + struct tc_fifo_qopt *ctl = nla_data(opt); + + if (nla_len(opt) < sizeof(*ctl)) + return -EINVAL; + + sch->limit = ctl->limit; + } + + if (is_bfifo) + bypass = sch->limit >= psched_mtu(qdisc_dev(sch)); + else + bypass = sch->limit >= 1; + + if (bypass) + sch->flags |= TCQ_F_CAN_BYPASS; + else + sch->flags &= ~TCQ_F_CAN_BYPASS; + return 0; +} + +static int fifo_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct tc_fifo_qopt opt = { .limit = sch->limit }; + + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + return -1; +} + +struct Qdisc_ops pfifo_qdisc_ops __read_mostly = { + .id = "pfifo", + .priv_size = 0, + .enqueue = pfifo_enqueue, + .dequeue = qdisc_dequeue_head, + .peek = qdisc_peek_head, + .init = fifo_init, + .reset = qdisc_reset_queue, + .change = fifo_init, + .dump = fifo_dump, + .owner = THIS_MODULE, +}; +EXPORT_SYMBOL(pfifo_qdisc_ops); + +struct Qdisc_ops bfifo_qdisc_ops __read_mostly = { + .id = "bfifo", + .priv_size = 0, + .enqueue = bfifo_enqueue, + .dequeue = qdisc_dequeue_head, + .peek = qdisc_peek_head, + .init = fifo_init, + .reset = qdisc_reset_queue, + .change = fifo_init, + .dump = fifo_dump, + .owner = THIS_MODULE, +}; +EXPORT_SYMBOL(bfifo_qdisc_ops); + +struct Qdisc_ops pfifo_head_drop_qdisc_ops __read_mostly = { + .id = "pfifo_head_drop", + .priv_size = 0, + .enqueue = pfifo_tail_enqueue, + .dequeue = qdisc_dequeue_head, + .peek = qdisc_peek_head, + .init = fifo_init, + .reset = qdisc_reset_queue, + .change = fifo_init, + .dump = fifo_dump, + .owner = THIS_MODULE, +}; + +/* Pass size change message down to embedded FIFO */ +int fifo_set_limit(struct Qdisc *q, unsigned int limit) +{ + struct nlattr *nla; + int ret = -ENOMEM; + + /* Hack to avoid sending change message to non-FIFO */ + if (strncmp(q->ops->id + 1, "fifo", 4) != 0) + return 0; + + if (!q->ops->change) + return 0; + + nla = kmalloc(nla_attr_size(sizeof(struct tc_fifo_qopt)), GFP_KERNEL); + if (nla) { + nla->nla_type = RTM_NEWQDISC; + nla->nla_len = nla_attr_size(sizeof(struct tc_fifo_qopt)); + ((struct tc_fifo_qopt *)nla_data(nla))->limit = limit; + + ret = q->ops->change(q, nla, NULL); + kfree(nla); + } + return ret; +} +EXPORT_SYMBOL(fifo_set_limit); + +struct Qdisc *fifo_create_dflt(struct Qdisc *sch, struct Qdisc_ops *ops, + unsigned int limit, + struct netlink_ext_ack *extack) +{ + struct Qdisc *q; + int err = -ENOMEM; + + q = qdisc_create_dflt(sch->dev_queue, ops, TC_H_MAKE(sch->handle, 1), + extack); + if (q) { + err = fifo_set_limit(q, limit); + if (err < 0) { + qdisc_put(q); + q = NULL; + } + } + + return q ? : ERR_PTR(err); +} +EXPORT_SYMBOL(fifo_create_dflt); diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c new file mode 100644 index 000000000..ba60a8dd5 --- /dev/null +++ b/net/sched/sch_fq.c @@ -0,0 +1,936 @@ +/* + * net/sched/sch_fq.c Fair Queue Packet Scheduler (per flow pacing) + * + * Copyright (C) 2013-2015 Eric Dumazet <edumazet@google.com> + * + * 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. + * + * Meant to be mostly used for locally generated traffic : + * Fast classification depends on skb->sk being set before reaching us. + * If not, (router workload), we use rxhash as fallback, with 32 bits wide hash. + * All packets belonging to a socket are considered as a 'flow'. + * + * Flows are dynamically allocated and stored in a hash table of RB trees + * They are also part of one Round Robin 'queues' (new or old flows) + * + * Burst avoidance (aka pacing) capability : + * + * Transport (eg TCP) can set in sk->sk_pacing_rate a rate, enqueue a + * bunch of packets, and this packet scheduler adds delay between + * packets to respect rate limitation. + * + * enqueue() : + * - lookup one RB tree (out of 1024 or more) to find the flow. + * If non existent flow, create it, add it to the tree. + * Add skb to the per flow list of skb (fifo). + * - Use a special fifo for high prio packets + * + * dequeue() : serves flows in Round Robin + * Note : When a flow becomes empty, we do not immediately remove it from + * rb trees, for performance reasons (its expected to send additional packets, + * or SLAB cache will reuse socket for another flow) + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/jiffies.h> +#include <linux/string.h> +#include <linux/in.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/skbuff.h> +#include <linux/slab.h> +#include <linux/rbtree.h> +#include <linux/hash.h> +#include <linux/prefetch.h> +#include <linux/vmalloc.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/sock.h> +#include <net/tcp_states.h> +#include <net/tcp.h> + +/* + * Per flow structure, dynamically allocated + */ +struct fq_flow { + struct sk_buff *head; /* list of skbs for this flow : first skb */ + union { + struct sk_buff *tail; /* last skb in the list */ + unsigned long age; /* jiffies when flow was emptied, for gc */ + }; + struct rb_node fq_node; /* anchor in fq_root[] trees */ + struct sock *sk; + int qlen; /* number of packets in flow queue */ + int credit; + u32 socket_hash; /* sk_hash */ + struct fq_flow *next; /* next pointer in RR lists, or &detached */ + + struct rb_node rate_node; /* anchor in q->delayed tree */ + u64 time_next_packet; +}; + +struct fq_flow_head { + struct fq_flow *first; + struct fq_flow *last; +}; + +struct fq_sched_data { + struct fq_flow_head new_flows; + + struct fq_flow_head old_flows; + + struct rb_root delayed; /* for rate limited flows */ + u64 time_next_delayed_flow; + unsigned long unthrottle_latency_ns; + + struct fq_flow internal; /* for non classified or high prio packets */ + u32 quantum; + u32 initial_quantum; + u32 flow_refill_delay; + u32 flow_max_rate; /* optional max rate per flow */ + u32 flow_plimit; /* max packets per flow */ + u32 orphan_mask; /* mask for orphaned skb */ + u32 low_rate_threshold; + struct rb_root *fq_root; + u8 rate_enable; + u8 fq_trees_log; + + u32 flows; + u32 inactive_flows; + u32 throttled_flows; + + u64 stat_gc_flows; + u64 stat_internal_packets; + u64 stat_tcp_retrans; + u64 stat_throttled; + u64 stat_flows_plimit; + u64 stat_pkts_too_long; + u64 stat_allocation_errors; + struct qdisc_watchdog watchdog; +}; + +/* special value to mark a detached flow (not on old/new list) */ +static struct fq_flow detached, throttled; + +static void fq_flow_set_detached(struct fq_flow *f) +{ + f->next = &detached; + f->age = jiffies; +} + +static bool fq_flow_is_detached(const struct fq_flow *f) +{ + return f->next == &detached; +} + +static bool fq_flow_is_throttled(const struct fq_flow *f) +{ + return f->next == &throttled; +} + +static void fq_flow_add_tail(struct fq_flow_head *head, struct fq_flow *flow) +{ + if (head->first) + head->last->next = flow; + else + head->first = flow; + head->last = flow; + flow->next = NULL; +} + +static void fq_flow_unset_throttled(struct fq_sched_data *q, struct fq_flow *f) +{ + rb_erase(&f->rate_node, &q->delayed); + q->throttled_flows--; + fq_flow_add_tail(&q->old_flows, f); +} + +static void fq_flow_set_throttled(struct fq_sched_data *q, struct fq_flow *f) +{ + struct rb_node **p = &q->delayed.rb_node, *parent = NULL; + + while (*p) { + struct fq_flow *aux; + + parent = *p; + aux = rb_entry(parent, struct fq_flow, rate_node); + if (f->time_next_packet >= aux->time_next_packet) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&f->rate_node, parent, p); + rb_insert_color(&f->rate_node, &q->delayed); + q->throttled_flows++; + q->stat_throttled++; + + f->next = &throttled; + if (q->time_next_delayed_flow > f->time_next_packet) + q->time_next_delayed_flow = f->time_next_packet; +} + + +static struct kmem_cache *fq_flow_cachep __read_mostly; + + +/* limit number of collected flows per round */ +#define FQ_GC_MAX 8 +#define FQ_GC_AGE (3*HZ) + +static bool fq_gc_candidate(const struct fq_flow *f) +{ + return fq_flow_is_detached(f) && + time_after(jiffies, f->age + FQ_GC_AGE); +} + +static void fq_gc(struct fq_sched_data *q, + struct rb_root *root, + struct sock *sk) +{ + struct fq_flow *f, *tofree[FQ_GC_MAX]; + struct rb_node **p, *parent; + int fcnt = 0; + + p = &root->rb_node; + parent = NULL; + while (*p) { + parent = *p; + + f = rb_entry(parent, struct fq_flow, fq_node); + if (f->sk == sk) + break; + + if (fq_gc_candidate(f)) { + tofree[fcnt++] = f; + if (fcnt == FQ_GC_MAX) + break; + } + + if (f->sk > sk) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + + q->flows -= fcnt; + q->inactive_flows -= fcnt; + q->stat_gc_flows += fcnt; + while (fcnt) { + struct fq_flow *f = tofree[--fcnt]; + + rb_erase(&f->fq_node, root); + kmem_cache_free(fq_flow_cachep, f); + } +} + +static struct fq_flow *fq_classify(struct sk_buff *skb, struct fq_sched_data *q) +{ + struct rb_node **p, *parent; + struct sock *sk = skb->sk; + struct rb_root *root; + struct fq_flow *f; + + /* warning: no starvation prevention... */ + if (unlikely((skb->priority & TC_PRIO_MAX) == TC_PRIO_CONTROL)) + return &q->internal; + + /* SYNACK messages are attached to a TCP_NEW_SYN_RECV request socket + * or a listener (SYNCOOKIE mode) + * 1) request sockets are not full blown, + * they do not contain sk_pacing_rate + * 2) They are not part of a 'flow' yet + * 3) We do not want to rate limit them (eg SYNFLOOD attack), + * especially if the listener set SO_MAX_PACING_RATE + * 4) We pretend they are orphaned + */ + if (!sk || sk_listener(sk)) { + unsigned long hash = skb_get_hash(skb) & q->orphan_mask; + + /* By forcing low order bit to 1, we make sure to not + * collide with a local flow (socket pointers are word aligned) + */ + sk = (struct sock *)((hash << 1) | 1UL); + skb_orphan(skb); + } + + root = &q->fq_root[hash_ptr(sk, q->fq_trees_log)]; + + if (q->flows >= (2U << q->fq_trees_log) && + q->inactive_flows > q->flows/2) + fq_gc(q, root, sk); + + p = &root->rb_node; + parent = NULL; + while (*p) { + parent = *p; + + f = rb_entry(parent, struct fq_flow, fq_node); + if (f->sk == sk) { + /* socket might have been reallocated, so check + * if its sk_hash is the same. + * It not, we need to refill credit with + * initial quantum + */ + if (unlikely(skb->sk && + f->socket_hash != sk->sk_hash)) { + f->credit = q->initial_quantum; + f->socket_hash = sk->sk_hash; + if (fq_flow_is_throttled(f)) + fq_flow_unset_throttled(q, f); + f->time_next_packet = 0ULL; + } + return f; + } + if (f->sk > sk) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + + f = kmem_cache_zalloc(fq_flow_cachep, GFP_ATOMIC | __GFP_NOWARN); + if (unlikely(!f)) { + q->stat_allocation_errors++; + return &q->internal; + } + fq_flow_set_detached(f); + f->sk = sk; + if (skb->sk) + f->socket_hash = sk->sk_hash; + f->credit = q->initial_quantum; + + rb_link_node(&f->fq_node, parent, p); + rb_insert_color(&f->fq_node, root); + + q->flows++; + q->inactive_flows++; + return f; +} + + +/* remove one skb from head of flow queue */ +static struct sk_buff *fq_dequeue_head(struct Qdisc *sch, struct fq_flow *flow) +{ + struct sk_buff *skb = flow->head; + + if (skb) { + flow->head = skb->next; + skb->next = NULL; + flow->qlen--; + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + } + return skb; +} + +/* We might add in the future detection of retransmits + * For the time being, just return false + */ +static bool skb_is_retransmit(struct sk_buff *skb) +{ + return false; +} + +/* add skb to flow queue + * flow queue is a linked list, kind of FIFO, except for TCP retransmits + * We special case tcp retransmits to be transmitted before other packets. + * We rely on fact that TCP retransmits are unlikely, so we do not waste + * a separate queue or a pointer. + * head-> [retrans pkt 1] + * [retrans pkt 2] + * [ normal pkt 1] + * [ normal pkt 2] + * [ normal pkt 3] + * tail-> [ normal pkt 4] + */ +static void flow_queue_add(struct fq_flow *flow, struct sk_buff *skb) +{ + struct sk_buff *prev, *head = flow->head; + + skb->next = NULL; + if (!head) { + flow->head = skb; + flow->tail = skb; + return; + } + if (likely(!skb_is_retransmit(skb))) { + flow->tail->next = skb; + flow->tail = skb; + return; + } + + /* This skb is a tcp retransmit, + * find the last retrans packet in the queue + */ + prev = NULL; + while (skb_is_retransmit(head)) { + prev = head; + head = head->next; + if (!head) + break; + } + if (!prev) { /* no rtx packet in queue, become the new head */ + skb->next = flow->head; + flow->head = skb; + } else { + if (prev == flow->tail) + flow->tail = skb; + else + skb->next = prev->next; + prev->next = skb; + } +} + +static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct fq_sched_data *q = qdisc_priv(sch); + struct fq_flow *f; + + if (unlikely(sch->q.qlen >= sch->limit)) + return qdisc_drop(skb, sch, to_free); + + f = fq_classify(skb, q); + if (unlikely(f->qlen >= q->flow_plimit && f != &q->internal)) { + q->stat_flows_plimit++; + return qdisc_drop(skb, sch, to_free); + } + + f->qlen++; + if (skb_is_retransmit(skb)) + q->stat_tcp_retrans++; + qdisc_qstats_backlog_inc(sch, skb); + if (fq_flow_is_detached(f)) { + struct sock *sk = skb->sk; + + fq_flow_add_tail(&q->new_flows, f); + if (time_after(jiffies, f->age + q->flow_refill_delay)) + f->credit = max_t(u32, f->credit, q->quantum); + if (sk && q->rate_enable) { + if (unlikely(smp_load_acquire(&sk->sk_pacing_status) != + SK_PACING_FQ)) + smp_store_release(&sk->sk_pacing_status, + SK_PACING_FQ); + } + q->inactive_flows--; + } + + /* Note: this overwrites f->age */ + flow_queue_add(f, skb); + + if (unlikely(f == &q->internal)) { + q->stat_internal_packets++; + } + sch->q.qlen++; + + return NET_XMIT_SUCCESS; +} + +static void fq_check_throttled(struct fq_sched_data *q, u64 now) +{ + unsigned long sample; + struct rb_node *p; + + if (q->time_next_delayed_flow > now) + return; + + /* Update unthrottle latency EWMA. + * This is cheap and can help diagnosing timer/latency problems. + */ + sample = (unsigned long)(now - q->time_next_delayed_flow); + q->unthrottle_latency_ns -= q->unthrottle_latency_ns >> 3; + q->unthrottle_latency_ns += sample >> 3; + + q->time_next_delayed_flow = ~0ULL; + while ((p = rb_first(&q->delayed)) != NULL) { + struct fq_flow *f = rb_entry(p, struct fq_flow, rate_node); + + if (f->time_next_packet > now) { + q->time_next_delayed_flow = f->time_next_packet; + break; + } + fq_flow_unset_throttled(q, f); + } +} + +static struct sk_buff *fq_dequeue(struct Qdisc *sch) +{ + struct fq_sched_data *q = qdisc_priv(sch); + u64 now = ktime_get_ns(); + struct fq_flow_head *head; + struct sk_buff *skb; + struct fq_flow *f; + u32 rate, plen; + + skb = fq_dequeue_head(sch, &q->internal); + if (skb) + goto out; + fq_check_throttled(q, now); +begin: + head = &q->new_flows; + if (!head->first) { + head = &q->old_flows; + if (!head->first) { + if (q->time_next_delayed_flow != ~0ULL) + qdisc_watchdog_schedule_ns(&q->watchdog, + q->time_next_delayed_flow); + return NULL; + } + } + f = head->first; + + if (f->credit <= 0) { + f->credit += q->quantum; + head->first = f->next; + fq_flow_add_tail(&q->old_flows, f); + goto begin; + } + + skb = f->head; + if (unlikely(skb && now < f->time_next_packet && + !skb_is_tcp_pure_ack(skb))) { + head->first = f->next; + fq_flow_set_throttled(q, f); + goto begin; + } + + skb = fq_dequeue_head(sch, f); + if (!skb) { + head->first = f->next; + /* force a pass through old_flows to prevent starvation */ + if ((head == &q->new_flows) && q->old_flows.first) { + fq_flow_add_tail(&q->old_flows, f); + } else { + fq_flow_set_detached(f); + q->inactive_flows++; + } + goto begin; + } + prefetch(&skb->end); + f->credit -= qdisc_pkt_len(skb); + + if (!q->rate_enable) + goto out; + + /* Do not pace locally generated ack packets */ + if (skb_is_tcp_pure_ack(skb)) + goto out; + + rate = q->flow_max_rate; + if (skb->sk) + rate = min(skb->sk->sk_pacing_rate, rate); + + if (rate <= q->low_rate_threshold) { + f->credit = 0; + plen = qdisc_pkt_len(skb); + } else { + plen = max(qdisc_pkt_len(skb), q->quantum); + if (f->credit > 0) + goto out; + } + if (rate != ~0U) { + u64 len = (u64)plen * NSEC_PER_SEC; + + if (likely(rate)) + do_div(len, rate); + /* Since socket rate can change later, + * clamp the delay to 1 second. + * Really, providers of too big packets should be fixed ! + */ + if (unlikely(len > NSEC_PER_SEC)) { + len = NSEC_PER_SEC; + q->stat_pkts_too_long++; + } + /* Account for schedule/timers drifts. + * f->time_next_packet was set when prior packet was sent, + * and current time (@now) can be too late by tens of us. + */ + if (f->time_next_packet) + len -= min(len/2, now - f->time_next_packet); + f->time_next_packet = now + len; + } +out: + qdisc_bstats_update(sch, skb); + return skb; +} + +static void fq_flow_purge(struct fq_flow *flow) +{ + rtnl_kfree_skbs(flow->head, flow->tail); + flow->head = NULL; + flow->qlen = 0; +} + +static void fq_reset(struct Qdisc *sch) +{ + struct fq_sched_data *q = qdisc_priv(sch); + struct rb_root *root; + struct rb_node *p; + struct fq_flow *f; + unsigned int idx; + + sch->q.qlen = 0; + sch->qstats.backlog = 0; + + fq_flow_purge(&q->internal); + + if (!q->fq_root) + return; + + for (idx = 0; idx < (1U << q->fq_trees_log); idx++) { + root = &q->fq_root[idx]; + while ((p = rb_first(root)) != NULL) { + f = rb_entry(p, struct fq_flow, fq_node); + rb_erase(p, root); + + fq_flow_purge(f); + + kmem_cache_free(fq_flow_cachep, f); + } + } + q->new_flows.first = NULL; + q->old_flows.first = NULL; + q->delayed = RB_ROOT; + q->flows = 0; + q->inactive_flows = 0; + q->throttled_flows = 0; +} + +static void fq_rehash(struct fq_sched_data *q, + struct rb_root *old_array, u32 old_log, + struct rb_root *new_array, u32 new_log) +{ + struct rb_node *op, **np, *parent; + struct rb_root *oroot, *nroot; + struct fq_flow *of, *nf; + int fcnt = 0; + u32 idx; + + for (idx = 0; idx < (1U << old_log); idx++) { + oroot = &old_array[idx]; + while ((op = rb_first(oroot)) != NULL) { + rb_erase(op, oroot); + of = rb_entry(op, struct fq_flow, fq_node); + if (fq_gc_candidate(of)) { + fcnt++; + kmem_cache_free(fq_flow_cachep, of); + continue; + } + nroot = &new_array[hash_ptr(of->sk, new_log)]; + + np = &nroot->rb_node; + parent = NULL; + while (*np) { + parent = *np; + + nf = rb_entry(parent, struct fq_flow, fq_node); + BUG_ON(nf->sk == of->sk); + + if (nf->sk > of->sk) + np = &parent->rb_right; + else + np = &parent->rb_left; + } + + rb_link_node(&of->fq_node, parent, np); + rb_insert_color(&of->fq_node, nroot); + } + } + q->flows -= fcnt; + q->inactive_flows -= fcnt; + q->stat_gc_flows += fcnt; +} + +static void fq_free(void *addr) +{ + kvfree(addr); +} + +static int fq_resize(struct Qdisc *sch, u32 log) +{ + struct fq_sched_data *q = qdisc_priv(sch); + struct rb_root *array; + void *old_fq_root; + u32 idx; + + if (q->fq_root && log == q->fq_trees_log) + return 0; + + /* If XPS was setup, we can allocate memory on right NUMA node */ + array = kvmalloc_node(sizeof(struct rb_root) << log, GFP_KERNEL | __GFP_RETRY_MAYFAIL, + netdev_queue_numa_node_read(sch->dev_queue)); + if (!array) + return -ENOMEM; + + for (idx = 0; idx < (1U << log); idx++) + array[idx] = RB_ROOT; + + sch_tree_lock(sch); + + old_fq_root = q->fq_root; + if (old_fq_root) + fq_rehash(q, old_fq_root, q->fq_trees_log, array, log); + + q->fq_root = array; + q->fq_trees_log = log; + + sch_tree_unlock(sch); + + fq_free(old_fq_root); + + return 0; +} + +static const struct nla_policy fq_policy[TCA_FQ_MAX + 1] = { + [TCA_FQ_PLIMIT] = { .type = NLA_U32 }, + [TCA_FQ_FLOW_PLIMIT] = { .type = NLA_U32 }, + [TCA_FQ_QUANTUM] = { .type = NLA_U32 }, + [TCA_FQ_INITIAL_QUANTUM] = { .type = NLA_U32 }, + [TCA_FQ_RATE_ENABLE] = { .type = NLA_U32 }, + [TCA_FQ_FLOW_DEFAULT_RATE] = { .type = NLA_U32 }, + [TCA_FQ_FLOW_MAX_RATE] = { .type = NLA_U32 }, + [TCA_FQ_BUCKETS_LOG] = { .type = NLA_U32 }, + [TCA_FQ_FLOW_REFILL_DELAY] = { .type = NLA_U32 }, + [TCA_FQ_ORPHAN_MASK] = { .type = NLA_U32 }, + [TCA_FQ_LOW_RATE_THRESHOLD] = { .type = NLA_U32 }, +}; + +static int fq_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct fq_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_FQ_MAX + 1]; + int err, drop_count = 0; + unsigned drop_len = 0; + u32 fq_log; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_FQ_MAX, opt, fq_policy, NULL); + if (err < 0) + return err; + + sch_tree_lock(sch); + + fq_log = q->fq_trees_log; + + if (tb[TCA_FQ_BUCKETS_LOG]) { + u32 nval = nla_get_u32(tb[TCA_FQ_BUCKETS_LOG]); + + if (nval >= 1 && nval <= ilog2(256*1024)) + fq_log = nval; + else + err = -EINVAL; + } + if (tb[TCA_FQ_PLIMIT]) + sch->limit = nla_get_u32(tb[TCA_FQ_PLIMIT]); + + if (tb[TCA_FQ_FLOW_PLIMIT]) + q->flow_plimit = nla_get_u32(tb[TCA_FQ_FLOW_PLIMIT]); + + if (tb[TCA_FQ_QUANTUM]) { + u32 quantum = nla_get_u32(tb[TCA_FQ_QUANTUM]); + + if (quantum > 0 && quantum <= (1 << 20)) { + q->quantum = quantum; + } else { + NL_SET_ERR_MSG_MOD(extack, "invalid quantum"); + err = -EINVAL; + } + } + + if (tb[TCA_FQ_INITIAL_QUANTUM]) + q->initial_quantum = nla_get_u32(tb[TCA_FQ_INITIAL_QUANTUM]); + + if (tb[TCA_FQ_FLOW_DEFAULT_RATE]) + pr_warn_ratelimited("sch_fq: defrate %u ignored.\n", + nla_get_u32(tb[TCA_FQ_FLOW_DEFAULT_RATE])); + + if (tb[TCA_FQ_FLOW_MAX_RATE]) + q->flow_max_rate = nla_get_u32(tb[TCA_FQ_FLOW_MAX_RATE]); + + if (tb[TCA_FQ_LOW_RATE_THRESHOLD]) + q->low_rate_threshold = + nla_get_u32(tb[TCA_FQ_LOW_RATE_THRESHOLD]); + + if (tb[TCA_FQ_RATE_ENABLE]) { + u32 enable = nla_get_u32(tb[TCA_FQ_RATE_ENABLE]); + + if (enable <= 1) + q->rate_enable = enable; + else + err = -EINVAL; + } + + if (tb[TCA_FQ_FLOW_REFILL_DELAY]) { + u32 usecs_delay = nla_get_u32(tb[TCA_FQ_FLOW_REFILL_DELAY]) ; + + q->flow_refill_delay = usecs_to_jiffies(usecs_delay); + } + + if (tb[TCA_FQ_ORPHAN_MASK]) + q->orphan_mask = nla_get_u32(tb[TCA_FQ_ORPHAN_MASK]); + + if (!err) { + sch_tree_unlock(sch); + err = fq_resize(sch, fq_log); + sch_tree_lock(sch); + } + while (sch->q.qlen > sch->limit) { + struct sk_buff *skb = fq_dequeue(sch); + + if (!skb) + break; + drop_len += qdisc_pkt_len(skb); + rtnl_kfree_skbs(skb, skb); + drop_count++; + } + qdisc_tree_reduce_backlog(sch, drop_count, drop_len); + + sch_tree_unlock(sch); + return err; +} + +static void fq_destroy(struct Qdisc *sch) +{ + struct fq_sched_data *q = qdisc_priv(sch); + + fq_reset(sch); + fq_free(q->fq_root); + qdisc_watchdog_cancel(&q->watchdog); +} + +static int fq_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct fq_sched_data *q = qdisc_priv(sch); + int err; + + sch->limit = 10000; + q->flow_plimit = 100; + q->quantum = 2 * psched_mtu(qdisc_dev(sch)); + q->initial_quantum = 10 * psched_mtu(qdisc_dev(sch)); + q->flow_refill_delay = msecs_to_jiffies(40); + q->flow_max_rate = ~0U; + q->time_next_delayed_flow = ~0ULL; + q->rate_enable = 1; + q->new_flows.first = NULL; + q->old_flows.first = NULL; + q->delayed = RB_ROOT; + q->fq_root = NULL; + q->fq_trees_log = ilog2(1024); + q->orphan_mask = 1024 - 1; + q->low_rate_threshold = 550000 / 8; + qdisc_watchdog_init(&q->watchdog, sch); + + if (opt) + err = fq_change(sch, opt, extack); + else + err = fq_resize(sch, q->fq_trees_log); + + return err; +} + +static int fq_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct fq_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + + /* TCA_FQ_FLOW_DEFAULT_RATE is not used anymore */ + + if (nla_put_u32(skb, TCA_FQ_PLIMIT, sch->limit) || + nla_put_u32(skb, TCA_FQ_FLOW_PLIMIT, q->flow_plimit) || + nla_put_u32(skb, TCA_FQ_QUANTUM, q->quantum) || + nla_put_u32(skb, TCA_FQ_INITIAL_QUANTUM, q->initial_quantum) || + nla_put_u32(skb, TCA_FQ_RATE_ENABLE, q->rate_enable) || + nla_put_u32(skb, TCA_FQ_FLOW_MAX_RATE, q->flow_max_rate) || + nla_put_u32(skb, TCA_FQ_FLOW_REFILL_DELAY, + jiffies_to_usecs(q->flow_refill_delay)) || + nla_put_u32(skb, TCA_FQ_ORPHAN_MASK, q->orphan_mask) || + nla_put_u32(skb, TCA_FQ_LOW_RATE_THRESHOLD, + q->low_rate_threshold) || + nla_put_u32(skb, TCA_FQ_BUCKETS_LOG, q->fq_trees_log)) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + return -1; +} + +static int fq_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct fq_sched_data *q = qdisc_priv(sch); + struct tc_fq_qd_stats st; + + sch_tree_lock(sch); + + st.gc_flows = q->stat_gc_flows; + st.highprio_packets = q->stat_internal_packets; + st.tcp_retrans = q->stat_tcp_retrans; + st.throttled = q->stat_throttled; + st.flows_plimit = q->stat_flows_plimit; + st.pkts_too_long = q->stat_pkts_too_long; + st.allocation_errors = q->stat_allocation_errors; + st.time_next_delayed_flow = q->time_next_delayed_flow - ktime_get_ns(); + st.flows = q->flows; + st.inactive_flows = q->inactive_flows; + st.throttled_flows = q->throttled_flows; + st.unthrottle_latency_ns = min_t(unsigned long, + q->unthrottle_latency_ns, ~0U); + sch_tree_unlock(sch); + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static struct Qdisc_ops fq_qdisc_ops __read_mostly = { + .id = "fq", + .priv_size = sizeof(struct fq_sched_data), + + .enqueue = fq_enqueue, + .dequeue = fq_dequeue, + .peek = qdisc_peek_dequeued, + .init = fq_init, + .reset = fq_reset, + .destroy = fq_destroy, + .change = fq_change, + .dump = fq_dump, + .dump_stats = fq_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init fq_module_init(void) +{ + int ret; + + fq_flow_cachep = kmem_cache_create("fq_flow_cache", + sizeof(struct fq_flow), + 0, 0, NULL); + if (!fq_flow_cachep) + return -ENOMEM; + + ret = register_qdisc(&fq_qdisc_ops); + if (ret) + kmem_cache_destroy(fq_flow_cachep); + return ret; +} + +static void __exit fq_module_exit(void) +{ + unregister_qdisc(&fq_qdisc_ops); + kmem_cache_destroy(fq_flow_cachep); +} + +module_init(fq_module_init) +module_exit(fq_module_exit) +MODULE_AUTHOR("Eric Dumazet"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c new file mode 100644 index 000000000..e4f69c779 --- /dev/null +++ b/net/sched/sch_fq_codel.c @@ -0,0 +1,744 @@ +/* + * Fair Queue CoDel discipline + * + * 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. + * + * Copyright (C) 2012,2015 Eric Dumazet <edumazet@google.com> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/jiffies.h> +#include <linux/string.h> +#include <linux/in.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/skbuff.h> +#include <linux/jhash.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/codel.h> +#include <net/codel_impl.h> +#include <net/codel_qdisc.h> + +/* Fair Queue CoDel. + * + * Principles : + * Packets are classified (internal classifier or external) on flows. + * This is a Stochastic model (as we use a hash, several flows + * might be hashed on same slot) + * Each flow has a CoDel managed queue. + * Flows are linked onto two (Round Robin) lists, + * so that new flows have priority on old ones. + * + * For a given flow, packets are not reordered (CoDel uses a FIFO) + * head drops only. + * ECN capability is on by default. + * Low memory footprint (64 bytes per flow) + */ + +struct fq_codel_flow { + struct sk_buff *head; + struct sk_buff *tail; + struct list_head flowchain; + int deficit; + u32 dropped; /* number of drops (or ECN marks) on this flow */ + struct codel_vars cvars; +}; /* please try to keep this structure <= 64 bytes */ + +struct fq_codel_sched_data { + struct tcf_proto __rcu *filter_list; /* optional external classifier */ + struct tcf_block *block; + struct fq_codel_flow *flows; /* Flows table [flows_cnt] */ + u32 *backlogs; /* backlog table [flows_cnt] */ + u32 flows_cnt; /* number of flows */ + u32 quantum; /* psched_mtu(qdisc_dev(sch)); */ + u32 drop_batch_size; + u32 memory_limit; + struct codel_params cparams; + struct codel_stats cstats; + u32 memory_usage; + u32 drop_overmemory; + u32 drop_overlimit; + u32 new_flow_count; + + struct list_head new_flows; /* list of new flows */ + struct list_head old_flows; /* list of old flows */ +}; + +static unsigned int fq_codel_hash(const struct fq_codel_sched_data *q, + struct sk_buff *skb) +{ + return reciprocal_scale(skb_get_hash(skb), q->flows_cnt); +} + +static unsigned int fq_codel_classify(struct sk_buff *skb, struct Qdisc *sch, + int *qerr) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct tcf_proto *filter; + struct tcf_result res; + int result; + + if (TC_H_MAJ(skb->priority) == sch->handle && + TC_H_MIN(skb->priority) > 0 && + TC_H_MIN(skb->priority) <= q->flows_cnt) + return TC_H_MIN(skb->priority); + + filter = rcu_dereference_bh(q->filter_list); + if (!filter) + return fq_codel_hash(q, skb) + 1; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + result = tcf_classify(skb, filter, &res, false); + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return 0; + } +#endif + if (TC_H_MIN(res.classid) <= q->flows_cnt) + return TC_H_MIN(res.classid); + } + return 0; +} + +/* helper functions : might be changed when/if skb use a standard list_head */ + +/* remove one skb from head of slot queue */ +static inline struct sk_buff *dequeue_head(struct fq_codel_flow *flow) +{ + struct sk_buff *skb = flow->head; + + flow->head = skb->next; + skb->next = NULL; + return skb; +} + +/* add skb to flow queue (tail add) */ +static inline void flow_queue_add(struct fq_codel_flow *flow, + struct sk_buff *skb) +{ + if (flow->head == NULL) + flow->head = skb; + else + flow->tail->next = skb; + flow->tail = skb; + skb->next = NULL; +} + +static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets, + struct sk_buff **to_free) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + unsigned int maxbacklog = 0, idx = 0, i, len; + struct fq_codel_flow *flow; + unsigned int threshold; + unsigned int mem = 0; + + /* Queue is full! Find the fat flow and drop packet(s) from it. + * This might sound expensive, but with 1024 flows, we scan + * 4KB of memory, and we dont need to handle a complex tree + * in fast path (packet queue/enqueue) with many cache misses. + * In stress mode, we'll try to drop 64 packets from the flow, + * amortizing this linear lookup to one cache line per drop. + */ + for (i = 0; i < q->flows_cnt; i++) { + if (q->backlogs[i] > maxbacklog) { + maxbacklog = q->backlogs[i]; + idx = i; + } + } + + /* Our goal is to drop half of this fat flow backlog */ + threshold = maxbacklog >> 1; + + flow = &q->flows[idx]; + len = 0; + i = 0; + do { + skb = dequeue_head(flow); + len += qdisc_pkt_len(skb); + mem += get_codel_cb(skb)->mem_usage; + __qdisc_drop(skb, to_free); + } while (++i < max_packets && len < threshold); + + flow->dropped += i; + q->backlogs[idx] -= len; + q->memory_usage -= mem; + sch->qstats.drops += i; + sch->qstats.backlog -= len; + sch->q.qlen -= i; + return idx; +} + +static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + unsigned int idx, prev_backlog, prev_qlen; + struct fq_codel_flow *flow; + int uninitialized_var(ret); + unsigned int pkt_len; + bool memory_limited; + + idx = fq_codel_classify(skb, sch, &ret); + if (idx == 0) { + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; + } + idx--; + + codel_set_enqueue_time(skb); + flow = &q->flows[idx]; + flow_queue_add(flow, skb); + q->backlogs[idx] += qdisc_pkt_len(skb); + qdisc_qstats_backlog_inc(sch, skb); + + if (list_empty(&flow->flowchain)) { + list_add_tail(&flow->flowchain, &q->new_flows); + q->new_flow_count++; + flow->deficit = q->quantum; + flow->dropped = 0; + } + get_codel_cb(skb)->mem_usage = skb->truesize; + q->memory_usage += get_codel_cb(skb)->mem_usage; + memory_limited = q->memory_usage > q->memory_limit; + if (++sch->q.qlen <= sch->limit && !memory_limited) + return NET_XMIT_SUCCESS; + + prev_backlog = sch->qstats.backlog; + prev_qlen = sch->q.qlen; + + /* save this packet length as it might be dropped by fq_codel_drop() */ + pkt_len = qdisc_pkt_len(skb); + /* fq_codel_drop() is quite expensive, as it performs a linear search + * in q->backlogs[] to find a fat flow. + * So instead of dropping a single packet, drop half of its backlog + * with a 64 packets limit to not add a too big cpu spike here. + */ + ret = fq_codel_drop(sch, q->drop_batch_size, to_free); + + prev_qlen -= sch->q.qlen; + prev_backlog -= sch->qstats.backlog; + q->drop_overlimit += prev_qlen; + if (memory_limited) + q->drop_overmemory += prev_qlen; + + /* As we dropped packet(s), better let upper stack know this. + * If we dropped a packet for this flow, return NET_XMIT_CN, + * but in this case, our parents wont increase their backlogs. + */ + if (ret == idx) { + qdisc_tree_reduce_backlog(sch, prev_qlen - 1, + prev_backlog - pkt_len); + return NET_XMIT_CN; + } + qdisc_tree_reduce_backlog(sch, prev_qlen, prev_backlog); + return NET_XMIT_SUCCESS; +} + +/* This is the specific function called from codel_dequeue() + * to dequeue a packet from queue. Note: backlog is handled in + * codel, we dont need to reduce it here. + */ +static struct sk_buff *dequeue_func(struct codel_vars *vars, void *ctx) +{ + struct Qdisc *sch = ctx; + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct fq_codel_flow *flow; + struct sk_buff *skb = NULL; + + flow = container_of(vars, struct fq_codel_flow, cvars); + if (flow->head) { + skb = dequeue_head(flow); + q->backlogs[flow - q->flows] -= qdisc_pkt_len(skb); + q->memory_usage -= get_codel_cb(skb)->mem_usage; + sch->q.qlen--; + sch->qstats.backlog -= qdisc_pkt_len(skb); + } + return skb; +} + +static void drop_func(struct sk_buff *skb, void *ctx) +{ + struct Qdisc *sch = ctx; + + kfree_skb(skb); + qdisc_qstats_drop(sch); +} + +static struct sk_buff *fq_codel_dequeue(struct Qdisc *sch) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + struct fq_codel_flow *flow; + struct list_head *head; + u32 prev_drop_count, prev_ecn_mark; + +begin: + head = &q->new_flows; + if (list_empty(head)) { + head = &q->old_flows; + if (list_empty(head)) + return NULL; + } + flow = list_first_entry(head, struct fq_codel_flow, flowchain); + + if (flow->deficit <= 0) { + flow->deficit += q->quantum; + list_move_tail(&flow->flowchain, &q->old_flows); + goto begin; + } + + prev_drop_count = q->cstats.drop_count; + prev_ecn_mark = q->cstats.ecn_mark; + + skb = codel_dequeue(sch, &sch->qstats.backlog, &q->cparams, + &flow->cvars, &q->cstats, qdisc_pkt_len, + codel_get_enqueue_time, drop_func, dequeue_func); + + flow->dropped += q->cstats.drop_count - prev_drop_count; + flow->dropped += q->cstats.ecn_mark - prev_ecn_mark; + + if (!skb) { + /* force a pass through old_flows to prevent starvation */ + if ((head == &q->new_flows) && !list_empty(&q->old_flows)) + list_move_tail(&flow->flowchain, &q->old_flows); + else + list_del_init(&flow->flowchain); + goto begin; + } + qdisc_bstats_update(sch, skb); + flow->deficit -= qdisc_pkt_len(skb); + /* We cant call qdisc_tree_reduce_backlog() if our qlen is 0, + * or HTB crashes. Defer it for next round. + */ + if (q->cstats.drop_count && sch->q.qlen) { + qdisc_tree_reduce_backlog(sch, q->cstats.drop_count, + q->cstats.drop_len); + q->cstats.drop_count = 0; + q->cstats.drop_len = 0; + } + return skb; +} + +static void fq_codel_flow_purge(struct fq_codel_flow *flow) +{ + rtnl_kfree_skbs(flow->head, flow->tail); + flow->head = NULL; +} + +static void fq_codel_reset(struct Qdisc *sch) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + int i; + + INIT_LIST_HEAD(&q->new_flows); + INIT_LIST_HEAD(&q->old_flows); + for (i = 0; i < q->flows_cnt; i++) { + struct fq_codel_flow *flow = q->flows + i; + + fq_codel_flow_purge(flow); + INIT_LIST_HEAD(&flow->flowchain); + codel_vars_init(&flow->cvars); + } + memset(q->backlogs, 0, q->flows_cnt * sizeof(u32)); + sch->q.qlen = 0; + sch->qstats.backlog = 0; + q->memory_usage = 0; +} + +static const struct nla_policy fq_codel_policy[TCA_FQ_CODEL_MAX + 1] = { + [TCA_FQ_CODEL_TARGET] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_LIMIT] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_INTERVAL] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_ECN] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_FLOWS] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_QUANTUM] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_CE_THRESHOLD] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_DROP_BATCH_SIZE] = { .type = NLA_U32 }, + [TCA_FQ_CODEL_MEMORY_LIMIT] = { .type = NLA_U32 }, +}; + +static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_FQ_CODEL_MAX + 1]; + u32 quantum = 0; + int err; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_FQ_CODEL_MAX, opt, fq_codel_policy, + NULL); + if (err < 0) + return err; + if (tb[TCA_FQ_CODEL_FLOWS]) { + if (q->flows) + return -EINVAL; + q->flows_cnt = nla_get_u32(tb[TCA_FQ_CODEL_FLOWS]); + if (!q->flows_cnt || + q->flows_cnt > 65536) + return -EINVAL; + } + if (tb[TCA_FQ_CODEL_QUANTUM]) { + quantum = max(256U, nla_get_u32(tb[TCA_FQ_CODEL_QUANTUM])); + if (quantum > FQ_CODEL_QUANTUM_MAX) { + NL_SET_ERR_MSG(extack, "Invalid quantum"); + return -EINVAL; + } + } + sch_tree_lock(sch); + + if (tb[TCA_FQ_CODEL_TARGET]) { + u64 target = nla_get_u32(tb[TCA_FQ_CODEL_TARGET]); + + q->cparams.target = (target * NSEC_PER_USEC) >> CODEL_SHIFT; + } + + if (tb[TCA_FQ_CODEL_CE_THRESHOLD]) { + u64 val = nla_get_u32(tb[TCA_FQ_CODEL_CE_THRESHOLD]); + + q->cparams.ce_threshold = (val * NSEC_PER_USEC) >> CODEL_SHIFT; + } + + if (tb[TCA_FQ_CODEL_INTERVAL]) { + u64 interval = nla_get_u32(tb[TCA_FQ_CODEL_INTERVAL]); + + q->cparams.interval = (interval * NSEC_PER_USEC) >> CODEL_SHIFT; + } + + if (tb[TCA_FQ_CODEL_LIMIT]) + sch->limit = nla_get_u32(tb[TCA_FQ_CODEL_LIMIT]); + + if (tb[TCA_FQ_CODEL_ECN]) + q->cparams.ecn = !!nla_get_u32(tb[TCA_FQ_CODEL_ECN]); + + if (quantum) + q->quantum = quantum; + + if (tb[TCA_FQ_CODEL_DROP_BATCH_SIZE]) + q->drop_batch_size = max(1U, nla_get_u32(tb[TCA_FQ_CODEL_DROP_BATCH_SIZE])); + + if (tb[TCA_FQ_CODEL_MEMORY_LIMIT]) + q->memory_limit = min(1U << 31, nla_get_u32(tb[TCA_FQ_CODEL_MEMORY_LIMIT])); + + while (sch->q.qlen > sch->limit || + q->memory_usage > q->memory_limit) { + struct sk_buff *skb = fq_codel_dequeue(sch); + + q->cstats.drop_len += qdisc_pkt_len(skb); + rtnl_kfree_skbs(skb, skb); + q->cstats.drop_count++; + } + qdisc_tree_reduce_backlog(sch, q->cstats.drop_count, q->cstats.drop_len); + q->cstats.drop_count = 0; + q->cstats.drop_len = 0; + + sch_tree_unlock(sch); + return 0; +} + +static void fq_codel_destroy(struct Qdisc *sch) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + + tcf_block_put(q->block); + kvfree(q->backlogs); + kvfree(q->flows); +} + +static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + int i; + int err; + + sch->limit = 10*1024; + q->flows_cnt = 1024; + q->memory_limit = 32 << 20; /* 32 MBytes */ + q->drop_batch_size = 64; + q->quantum = psched_mtu(qdisc_dev(sch)); + INIT_LIST_HEAD(&q->new_flows); + INIT_LIST_HEAD(&q->old_flows); + codel_params_init(&q->cparams); + codel_stats_init(&q->cstats); + q->cparams.ecn = true; + q->cparams.mtu = psched_mtu(qdisc_dev(sch)); + + if (opt) { + err = fq_codel_change(sch, opt, extack); + if (err) + goto init_failure; + } + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + goto init_failure; + + if (!q->flows) { + q->flows = kvcalloc(q->flows_cnt, + sizeof(struct fq_codel_flow), + GFP_KERNEL); + if (!q->flows) { + err = -ENOMEM; + goto init_failure; + } + q->backlogs = kvcalloc(q->flows_cnt, sizeof(u32), GFP_KERNEL); + if (!q->backlogs) { + err = -ENOMEM; + goto alloc_failure; + } + for (i = 0; i < q->flows_cnt; i++) { + struct fq_codel_flow *flow = q->flows + i; + + INIT_LIST_HEAD(&flow->flowchain); + codel_vars_init(&flow->cvars); + } + } + if (sch->limit >= 1) + sch->flags |= TCQ_F_CAN_BYPASS; + else + sch->flags &= ~TCQ_F_CAN_BYPASS; + return 0; + +alloc_failure: + kvfree(q->flows); + q->flows = NULL; +init_failure: + q->flows_cnt = 0; + return err; +} + +static int fq_codel_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_FQ_CODEL_TARGET, + codel_time_to_us(q->cparams.target)) || + nla_put_u32(skb, TCA_FQ_CODEL_LIMIT, + sch->limit) || + nla_put_u32(skb, TCA_FQ_CODEL_INTERVAL, + codel_time_to_us(q->cparams.interval)) || + nla_put_u32(skb, TCA_FQ_CODEL_ECN, + q->cparams.ecn) || + nla_put_u32(skb, TCA_FQ_CODEL_QUANTUM, + q->quantum) || + nla_put_u32(skb, TCA_FQ_CODEL_DROP_BATCH_SIZE, + q->drop_batch_size) || + nla_put_u32(skb, TCA_FQ_CODEL_MEMORY_LIMIT, + q->memory_limit) || + nla_put_u32(skb, TCA_FQ_CODEL_FLOWS, + q->flows_cnt)) + goto nla_put_failure; + + if (q->cparams.ce_threshold != CODEL_DISABLED_THRESHOLD && + nla_put_u32(skb, TCA_FQ_CODEL_CE_THRESHOLD, + codel_time_to_us(q->cparams.ce_threshold))) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + return -1; +} + +static int fq_codel_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + struct tc_fq_codel_xstats st = { + .type = TCA_FQ_CODEL_XSTATS_QDISC, + }; + struct list_head *pos; + + st.qdisc_stats.maxpacket = q->cstats.maxpacket; + st.qdisc_stats.drop_overlimit = q->drop_overlimit; + st.qdisc_stats.ecn_mark = q->cstats.ecn_mark; + st.qdisc_stats.new_flow_count = q->new_flow_count; + st.qdisc_stats.ce_mark = q->cstats.ce_mark; + st.qdisc_stats.memory_usage = q->memory_usage; + st.qdisc_stats.drop_overmemory = q->drop_overmemory; + + sch_tree_lock(sch); + list_for_each(pos, &q->new_flows) + st.qdisc_stats.new_flows_len++; + + list_for_each(pos, &q->old_flows) + st.qdisc_stats.old_flows_len++; + sch_tree_unlock(sch); + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static struct Qdisc *fq_codel_leaf(struct Qdisc *sch, unsigned long arg) +{ + return NULL; +} + +static unsigned long fq_codel_find(struct Qdisc *sch, u32 classid) +{ + return 0; +} + +static unsigned long fq_codel_bind(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + return 0; +} + +static void fq_codel_unbind(struct Qdisc *q, unsigned long cl) +{ +} + +static struct tcf_block *fq_codel_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + + if (cl) + return NULL; + return q->block; +} + +static int fq_codel_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + tcm->tcm_handle |= TC_H_MIN(cl); + return 0; +} + +static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + u32 idx = cl - 1; + struct gnet_stats_queue qs = { 0 }; + struct tc_fq_codel_xstats xstats; + + if (idx < q->flows_cnt) { + const struct fq_codel_flow *flow = &q->flows[idx]; + const struct sk_buff *skb; + + memset(&xstats, 0, sizeof(xstats)); + xstats.type = TCA_FQ_CODEL_XSTATS_CLASS; + xstats.class_stats.deficit = flow->deficit; + xstats.class_stats.ldelay = + codel_time_to_us(flow->cvars.ldelay); + xstats.class_stats.count = flow->cvars.count; + xstats.class_stats.lastcount = flow->cvars.lastcount; + xstats.class_stats.dropping = flow->cvars.dropping; + if (flow->cvars.dropping) { + codel_tdiff_t delta = flow->cvars.drop_next - + codel_get_time(); + + xstats.class_stats.drop_next = (delta >= 0) ? + codel_time_to_us(delta) : + -codel_time_to_us(-delta); + } + if (flow->head) { + sch_tree_lock(sch); + skb = flow->head; + while (skb) { + qs.qlen++; + skb = skb->next; + } + sch_tree_unlock(sch); + } + qs.backlog = q->backlogs[idx]; + qs.drops = flow->dropped; + } + if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0) + return -1; + if (idx < q->flows_cnt) + return gnet_stats_copy_app(d, &xstats, sizeof(xstats)); + return 0; +} + +static void fq_codel_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct fq_codel_sched_data *q = qdisc_priv(sch); + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->flows_cnt; i++) { + if (list_empty(&q->flows[i].flowchain) || + arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, i + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } +} + +static const struct Qdisc_class_ops fq_codel_class_ops = { + .leaf = fq_codel_leaf, + .find = fq_codel_find, + .tcf_block = fq_codel_tcf_block, + .bind_tcf = fq_codel_bind, + .unbind_tcf = fq_codel_unbind, + .dump = fq_codel_dump_class, + .dump_stats = fq_codel_dump_class_stats, + .walk = fq_codel_walk, +}; + +static struct Qdisc_ops fq_codel_qdisc_ops __read_mostly = { + .cl_ops = &fq_codel_class_ops, + .id = "fq_codel", + .priv_size = sizeof(struct fq_codel_sched_data), + .enqueue = fq_codel_enqueue, + .dequeue = fq_codel_dequeue, + .peek = qdisc_peek_dequeued, + .init = fq_codel_init, + .reset = fq_codel_reset, + .destroy = fq_codel_destroy, + .change = fq_codel_change, + .dump = fq_codel_dump, + .dump_stats = fq_codel_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init fq_codel_module_init(void) +{ + return register_qdisc(&fq_codel_qdisc_ops); +} + +static void __exit fq_codel_module_exit(void) +{ + unregister_qdisc(&fq_codel_qdisc_ops); +} + +module_init(fq_codel_module_init) +module_exit(fq_codel_module_exit) +MODULE_AUTHOR("Eric Dumazet"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c new file mode 100644 index 000000000..7c1b1eff8 --- /dev/null +++ b/net/sched/sch_generic.c @@ -0,0 +1,1447 @@ +/* + * net/sched/sch_generic.c Generic packet scheduler routines. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * Jamal Hadi Salim, <hadi@cyberus.ca> 990601 + * - Ingress support + */ + +#include <linux/bitops.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/init.h> +#include <linux/rcupdate.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/if_vlan.h> +#include <linux/skb_array.h> +#include <linux/if_macvlan.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> +#include <net/dst.h> +#include <trace/events/qdisc.h> +#include <net/xfrm.h> + +/* Qdisc to use by default */ +const struct Qdisc_ops *default_qdisc_ops = &pfifo_fast_ops; +EXPORT_SYMBOL(default_qdisc_ops); + +/* Main transmission queue. */ + +/* Modifications to data participating in scheduling must be protected with + * qdisc_lock(qdisc) spinlock. + * + * The idea is the following: + * - enqueue, dequeue are serialized via qdisc root lock + * - ingress filtering is also serialized via qdisc root lock + * - updates to tree and tree walking are only done under the rtnl mutex. + */ + +#define SKB_XOFF_MAGIC ((struct sk_buff *)1UL) + +static inline struct sk_buff *__skb_dequeue_bad_txq(struct Qdisc *q) +{ + const struct netdev_queue *txq = q->dev_queue; + spinlock_t *lock = NULL; + struct sk_buff *skb; + + if (q->flags & TCQ_F_NOLOCK) { + lock = qdisc_lock(q); + spin_lock(lock); + } + + skb = skb_peek(&q->skb_bad_txq); + if (skb) { + /* check the reason of requeuing without tx lock first */ + txq = skb_get_tx_queue(txq->dev, skb); + if (!netif_xmit_frozen_or_stopped(txq)) { + skb = __skb_dequeue(&q->skb_bad_txq); + if (qdisc_is_percpu_stats(q)) { + qdisc_qstats_cpu_backlog_dec(q, skb); + qdisc_qstats_atomic_qlen_dec(q); + } else { + qdisc_qstats_backlog_dec(q, skb); + q->q.qlen--; + } + } else { + skb = SKB_XOFF_MAGIC; + } + } + + if (lock) + spin_unlock(lock); + + return skb; +} + +static inline struct sk_buff *qdisc_dequeue_skb_bad_txq(struct Qdisc *q) +{ + struct sk_buff *skb = skb_peek(&q->skb_bad_txq); + + if (unlikely(skb)) + skb = __skb_dequeue_bad_txq(q); + + return skb; +} + +static inline void qdisc_enqueue_skb_bad_txq(struct Qdisc *q, + struct sk_buff *skb) +{ + spinlock_t *lock = NULL; + + if (q->flags & TCQ_F_NOLOCK) { + lock = qdisc_lock(q); + spin_lock(lock); + } + + __skb_queue_tail(&q->skb_bad_txq, skb); + + if (qdisc_is_percpu_stats(q)) { + qdisc_qstats_cpu_backlog_inc(q, skb); + qdisc_qstats_atomic_qlen_inc(q); + } else { + qdisc_qstats_backlog_inc(q, skb); + q->q.qlen++; + } + + if (lock) + spin_unlock(lock); +} + +static inline int __dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q) +{ + while (skb) { + struct sk_buff *next = skb->next; + + __skb_queue_tail(&q->gso_skb, skb); + q->qstats.requeues++; + qdisc_qstats_backlog_inc(q, skb); + q->q.qlen++; /* it's still part of the queue */ + + skb = next; + } + __netif_schedule(q); + + return 0; +} + +static inline int dev_requeue_skb_locked(struct sk_buff *skb, struct Qdisc *q) +{ + spinlock_t *lock = qdisc_lock(q); + + spin_lock(lock); + while (skb) { + struct sk_buff *next = skb->next; + + __skb_queue_tail(&q->gso_skb, skb); + + qdisc_qstats_cpu_requeues_inc(q); + qdisc_qstats_cpu_backlog_inc(q, skb); + qdisc_qstats_atomic_qlen_inc(q); + + skb = next; + } + spin_unlock(lock); + + __netif_schedule(q); + + return 0; +} + +static inline int dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q) +{ + if (q->flags & TCQ_F_NOLOCK) + return dev_requeue_skb_locked(skb, q); + else + return __dev_requeue_skb(skb, q); +} + +static void try_bulk_dequeue_skb(struct Qdisc *q, + struct sk_buff *skb, + const struct netdev_queue *txq, + int *packets) +{ + int bytelimit = qdisc_avail_bulklimit(txq) - skb->len; + + while (bytelimit > 0) { + struct sk_buff *nskb = q->dequeue(q); + + if (!nskb) + break; + + bytelimit -= nskb->len; /* covers GSO len */ + skb->next = nskb; + skb = nskb; + (*packets)++; /* GSO counts as one pkt */ + } + skb->next = NULL; +} + +/* This variant of try_bulk_dequeue_skb() makes sure + * all skbs in the chain are for the same txq + */ +static void try_bulk_dequeue_skb_slow(struct Qdisc *q, + struct sk_buff *skb, + int *packets) +{ + int mapping = skb_get_queue_mapping(skb); + struct sk_buff *nskb; + int cnt = 0; + + do { + nskb = q->dequeue(q); + if (!nskb) + break; + if (unlikely(skb_get_queue_mapping(nskb) != mapping)) { + qdisc_enqueue_skb_bad_txq(q, nskb); + break; + } + skb->next = nskb; + skb = nskb; + } while (++cnt < 8); + (*packets) += cnt; + skb->next = NULL; +} + +/* Note that dequeue_skb can possibly return a SKB list (via skb->next). + * A requeued skb (via q->gso_skb) can also be a SKB list. + */ +static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, + int *packets) +{ + const struct netdev_queue *txq = q->dev_queue; + struct sk_buff *skb = NULL; + + *packets = 1; + if (unlikely(!skb_queue_empty(&q->gso_skb))) { + spinlock_t *lock = NULL; + + if (q->flags & TCQ_F_NOLOCK) { + lock = qdisc_lock(q); + spin_lock(lock); + } + + skb = skb_peek(&q->gso_skb); + + /* skb may be null if another cpu pulls gso_skb off in between + * empty check and lock. + */ + if (!skb) { + if (lock) + spin_unlock(lock); + goto validate; + } + + /* skb in gso_skb were already validated */ + *validate = false; + if (xfrm_offload(skb)) + *validate = true; + /* check the reason of requeuing without tx lock first */ + txq = skb_get_tx_queue(txq->dev, skb); + if (!netif_xmit_frozen_or_stopped(txq)) { + skb = __skb_dequeue(&q->gso_skb); + if (qdisc_is_percpu_stats(q)) { + qdisc_qstats_cpu_backlog_dec(q, skb); + qdisc_qstats_atomic_qlen_dec(q); + } else { + qdisc_qstats_backlog_dec(q, skb); + q->q.qlen--; + } + } else { + skb = NULL; + } + if (lock) + spin_unlock(lock); + goto trace; + } +validate: + *validate = true; + + if ((q->flags & TCQ_F_ONETXQUEUE) && + netif_xmit_frozen_or_stopped(txq)) + return skb; + + skb = qdisc_dequeue_skb_bad_txq(q); + if (unlikely(skb)) { + if (skb == SKB_XOFF_MAGIC) + return NULL; + goto bulk; + } + skb = q->dequeue(q); + if (skb) { +bulk: + if (qdisc_may_bulk(q)) + try_bulk_dequeue_skb(q, skb, txq, packets); + else + try_bulk_dequeue_skb_slow(q, skb, packets); + } +trace: + trace_qdisc_dequeue(q, txq, *packets, skb); + return skb; +} + +/* + * Transmit possibly several skbs, and handle the return status as + * required. Owning running seqcount bit guarantees that + * only one CPU can execute this function. + * + * Returns to the caller: + * false - hardware queue frozen backoff + * true - feel free to send more pkts + */ +bool sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, + struct net_device *dev, struct netdev_queue *txq, + spinlock_t *root_lock, bool validate) +{ + int ret = NETDEV_TX_BUSY; + bool again = false; + + /* And release qdisc */ + if (root_lock) + spin_unlock(root_lock); + + /* Note that we validate skb (GSO, checksum, ...) outside of locks */ + if (validate) + skb = validate_xmit_skb_list(skb, dev, &again); + +#ifdef CONFIG_XFRM_OFFLOAD + if (unlikely(again)) { + if (root_lock) + spin_lock(root_lock); + + dev_requeue_skb(skb, q); + return false; + } +#endif + + if (likely(skb)) { + HARD_TX_LOCK(dev, txq, smp_processor_id()); + if (!netif_xmit_frozen_or_stopped(txq)) + skb = dev_hard_start_xmit(skb, dev, txq, &ret); + + HARD_TX_UNLOCK(dev, txq); + } else { + if (root_lock) + spin_lock(root_lock); + return true; + } + + if (root_lock) + spin_lock(root_lock); + + if (!dev_xmit_complete(ret)) { + /* Driver returned NETDEV_TX_BUSY - requeue skb */ + if (unlikely(ret != NETDEV_TX_BUSY)) + net_warn_ratelimited("BUG %s code %d qlen %d\n", + dev->name, ret, q->q.qlen); + + dev_requeue_skb(skb, q); + return false; + } + + return true; +} + +/* + * NOTE: Called under qdisc_lock(q) with locally disabled BH. + * + * running seqcount guarantees only one CPU can process + * this qdisc at a time. qdisc_lock(q) serializes queue accesses for + * this queue. + * + * netif_tx_lock serializes accesses to device driver. + * + * qdisc_lock(q) and netif_tx_lock are mutually exclusive, + * if one is grabbed, another must be free. + * + * Note, that this procedure can be called by a watchdog timer + * + * Returns to the caller: + * 0 - queue is empty or throttled. + * >0 - queue is not empty. + * + */ +static inline bool qdisc_restart(struct Qdisc *q, int *packets) +{ + spinlock_t *root_lock = NULL; + struct netdev_queue *txq; + struct net_device *dev; + struct sk_buff *skb; + bool validate; + + /* Dequeue packet */ + skb = dequeue_skb(q, &validate, packets); + if (unlikely(!skb)) + return false; + + if (!(q->flags & TCQ_F_NOLOCK)) + root_lock = qdisc_lock(q); + + dev = qdisc_dev(q); + txq = skb_get_tx_queue(dev, skb); + + return sch_direct_xmit(skb, q, dev, txq, root_lock, validate); +} + +void __qdisc_run(struct Qdisc *q) +{ + int quota = dev_tx_weight; + int packets; + + while (qdisc_restart(q, &packets)) { + /* + * Ordered by possible occurrence: Postpone processing if + * 1. we've exceeded packet quota + * 2. another process needs the CPU; + */ + quota -= packets; + if (quota <= 0 || need_resched()) { + __netif_schedule(q); + break; + } + } +} + +unsigned long dev_trans_start(struct net_device *dev) +{ + unsigned long val, res; + unsigned int i; + + if (is_vlan_dev(dev)) + dev = vlan_dev_real_dev(dev); + else if (netif_is_macvlan(dev)) + dev = macvlan_dev_real_dev(dev); + res = netdev_get_tx_queue(dev, 0)->trans_start; + for (i = 1; i < dev->num_tx_queues; i++) { + val = netdev_get_tx_queue(dev, i)->trans_start; + if (val && time_after(val, res)) + res = val; + } + + return res; +} +EXPORT_SYMBOL(dev_trans_start); + +static void dev_watchdog(struct timer_list *t) +{ + struct net_device *dev = from_timer(dev, t, watchdog_timer); + + netif_tx_lock(dev); + if (!qdisc_tx_is_noop(dev)) { + if (netif_device_present(dev) && + netif_running(dev) && + netif_carrier_ok(dev)) { + int some_queue_timedout = 0; + unsigned int i; + unsigned long trans_start; + + for (i = 0; i < dev->num_tx_queues; i++) { + struct netdev_queue *txq; + + txq = netdev_get_tx_queue(dev, i); + trans_start = txq->trans_start; + if (netif_xmit_stopped(txq) && + time_after(jiffies, (trans_start + + dev->watchdog_timeo))) { + some_queue_timedout = 1; + txq->trans_timeout++; + break; + } + } + + if (some_queue_timedout) { + WARN_ONCE(1, KERN_INFO "NETDEV WATCHDOG: %s (%s): transmit queue %u timed out\n", + dev->name, netdev_drivername(dev), i); + dev->netdev_ops->ndo_tx_timeout(dev); + } + if (!mod_timer(&dev->watchdog_timer, + round_jiffies(jiffies + + dev->watchdog_timeo))) + dev_hold(dev); + } + } + netif_tx_unlock(dev); + + dev_put(dev); +} + +void __netdev_watchdog_up(struct net_device *dev) +{ + if (dev->netdev_ops->ndo_tx_timeout) { + if (dev->watchdog_timeo <= 0) + dev->watchdog_timeo = 5*HZ; + if (!mod_timer(&dev->watchdog_timer, + round_jiffies(jiffies + dev->watchdog_timeo))) + dev_hold(dev); + } +} +EXPORT_SYMBOL_GPL(__netdev_watchdog_up); + +static void dev_watchdog_up(struct net_device *dev) +{ + __netdev_watchdog_up(dev); +} + +static void dev_watchdog_down(struct net_device *dev) +{ + netif_tx_lock_bh(dev); + if (del_timer(&dev->watchdog_timer)) + dev_put(dev); + netif_tx_unlock_bh(dev); +} + +/** + * netif_carrier_on - set carrier + * @dev: network device + * + * Device has detected that carrier. + */ +void netif_carrier_on(struct net_device *dev) +{ + if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) { + if (dev->reg_state == NETREG_UNINITIALIZED) + return; + atomic_inc(&dev->carrier_up_count); + linkwatch_fire_event(dev); + if (netif_running(dev)) + __netdev_watchdog_up(dev); + } +} +EXPORT_SYMBOL(netif_carrier_on); + +/** + * netif_carrier_off - clear carrier + * @dev: network device + * + * Device has detected loss of carrier. + */ +void netif_carrier_off(struct net_device *dev) +{ + if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) { + if (dev->reg_state == NETREG_UNINITIALIZED) + return; + atomic_inc(&dev->carrier_down_count); + linkwatch_fire_event(dev); + } +} +EXPORT_SYMBOL(netif_carrier_off); + +/* "NOOP" scheduler: the best scheduler, recommended for all interfaces + under all circumstances. It is difficult to invent anything faster or + cheaper. + */ + +static int noop_enqueue(struct sk_buff *skb, struct Qdisc *qdisc, + struct sk_buff **to_free) +{ + __qdisc_drop(skb, to_free); + return NET_XMIT_CN; +} + +static struct sk_buff *noop_dequeue(struct Qdisc *qdisc) +{ + return NULL; +} + +struct Qdisc_ops noop_qdisc_ops __read_mostly = { + .id = "noop", + .priv_size = 0, + .enqueue = noop_enqueue, + .dequeue = noop_dequeue, + .peek = noop_dequeue, + .owner = THIS_MODULE, +}; + +static struct netdev_queue noop_netdev_queue = { + .qdisc = &noop_qdisc, + .qdisc_sleeping = &noop_qdisc, +}; + +struct Qdisc noop_qdisc = { + .enqueue = noop_enqueue, + .dequeue = noop_dequeue, + .flags = TCQ_F_BUILTIN, + .ops = &noop_qdisc_ops, + .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), + .dev_queue = &noop_netdev_queue, + .running = SEQCNT_ZERO(noop_qdisc.running), + .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock), + .gso_skb = { + .next = (struct sk_buff *)&noop_qdisc.gso_skb, + .prev = (struct sk_buff *)&noop_qdisc.gso_skb, + .qlen = 0, + .lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.gso_skb.lock), + }, + .skb_bad_txq = { + .next = (struct sk_buff *)&noop_qdisc.skb_bad_txq, + .prev = (struct sk_buff *)&noop_qdisc.skb_bad_txq, + .qlen = 0, + .lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.skb_bad_txq.lock), + }, +}; +EXPORT_SYMBOL(noop_qdisc); + +static int noqueue_init(struct Qdisc *qdisc, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + /* register_qdisc() assigns a default of noop_enqueue if unset, + * but __dev_queue_xmit() treats noqueue only as such + * if this is NULL - so clear it here. */ + qdisc->enqueue = NULL; + return 0; +} + +struct Qdisc_ops noqueue_qdisc_ops __read_mostly = { + .id = "noqueue", + .priv_size = 0, + .init = noqueue_init, + .enqueue = noop_enqueue, + .dequeue = noop_dequeue, + .peek = noop_dequeue, + .owner = THIS_MODULE, +}; + +static const u8 prio2band[TC_PRIO_MAX + 1] = { + 1, 2, 2, 2, 1, 2, 0, 0 , 1, 1, 1, 1, 1, 1, 1, 1 +}; + +/* 3-band FIFO queue: old style, but should be a bit faster than + generic prio+fifo combination. + */ + +#define PFIFO_FAST_BANDS 3 + +/* + * Private data for a pfifo_fast scheduler containing: + * - rings for priority bands + */ +struct pfifo_fast_priv { + struct skb_array q[PFIFO_FAST_BANDS]; +}; + +static inline struct skb_array *band2list(struct pfifo_fast_priv *priv, + int band) +{ + return &priv->q[band]; +} + +static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc, + struct sk_buff **to_free) +{ + int band = prio2band[skb->priority & TC_PRIO_MAX]; + struct pfifo_fast_priv *priv = qdisc_priv(qdisc); + struct skb_array *q = band2list(priv, band); + unsigned int pkt_len = qdisc_pkt_len(skb); + int err; + + err = skb_array_produce(q, skb); + + if (unlikely(err)) + return qdisc_drop_cpu(skb, qdisc, to_free); + + qdisc_qstats_atomic_qlen_inc(qdisc); + /* Note: skb can not be used after skb_array_produce(), + * so we better not use qdisc_qstats_cpu_backlog_inc() + */ + this_cpu_add(qdisc->cpu_qstats->backlog, pkt_len); + return NET_XMIT_SUCCESS; +} + +static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc) +{ + struct pfifo_fast_priv *priv = qdisc_priv(qdisc); + struct sk_buff *skb = NULL; + int band; + + for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) { + struct skb_array *q = band2list(priv, band); + + if (__skb_array_empty(q)) + continue; + + skb = __skb_array_consume(q); + } + if (likely(skb)) { + qdisc_qstats_cpu_backlog_dec(qdisc, skb); + qdisc_bstats_cpu_update(qdisc, skb); + qdisc_qstats_atomic_qlen_dec(qdisc); + } + + return skb; +} + +static struct sk_buff *pfifo_fast_peek(struct Qdisc *qdisc) +{ + struct pfifo_fast_priv *priv = qdisc_priv(qdisc); + struct sk_buff *skb = NULL; + int band; + + for (band = 0; band < PFIFO_FAST_BANDS && !skb; band++) { + struct skb_array *q = band2list(priv, band); + + skb = __skb_array_peek(q); + } + + return skb; +} + +static void pfifo_fast_reset(struct Qdisc *qdisc) +{ + int i, band; + struct pfifo_fast_priv *priv = qdisc_priv(qdisc); + + for (band = 0; band < PFIFO_FAST_BANDS; band++) { + struct skb_array *q = band2list(priv, band); + struct sk_buff *skb; + + /* NULL ring is possible if destroy path is due to a failed + * skb_array_init() in pfifo_fast_init() case. + */ + if (!q->ring.queue) + continue; + + while ((skb = __skb_array_consume(q)) != NULL) + kfree_skb(skb); + } + + for_each_possible_cpu(i) { + struct gnet_stats_queue *q = per_cpu_ptr(qdisc->cpu_qstats, i); + + q->backlog = 0; + } +} + +static int pfifo_fast_dump(struct Qdisc *qdisc, struct sk_buff *skb) +{ + struct tc_prio_qopt opt = { .bands = PFIFO_FAST_BANDS }; + + memcpy(&opt.priomap, prio2band, TC_PRIO_MAX + 1); + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + goto nla_put_failure; + return skb->len; + +nla_put_failure: + return -1; +} + +static int pfifo_fast_init(struct Qdisc *qdisc, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + unsigned int qlen = qdisc_dev(qdisc)->tx_queue_len; + struct pfifo_fast_priv *priv = qdisc_priv(qdisc); + int prio; + + /* guard against zero length rings */ + if (!qlen) + return -EINVAL; + + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) { + struct skb_array *q = band2list(priv, prio); + int err; + + err = skb_array_init(q, qlen, GFP_KERNEL); + if (err) + return -ENOMEM; + } + + /* Can by-pass the queue discipline */ + qdisc->flags |= TCQ_F_CAN_BYPASS; + return 0; +} + +static void pfifo_fast_destroy(struct Qdisc *sch) +{ + struct pfifo_fast_priv *priv = qdisc_priv(sch); + int prio; + + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) { + struct skb_array *q = band2list(priv, prio); + + /* NULL ring is possible if destroy path is due to a failed + * skb_array_init() in pfifo_fast_init() case. + */ + if (!q->ring.queue) + continue; + /* Destroy ring but no need to kfree_skb because a call to + * pfifo_fast_reset() has already done that work. + */ + ptr_ring_cleanup(&q->ring, NULL); + } +} + +static int pfifo_fast_change_tx_queue_len(struct Qdisc *sch, + unsigned int new_len) +{ + struct pfifo_fast_priv *priv = qdisc_priv(sch); + struct skb_array *bands[PFIFO_FAST_BANDS]; + int prio; + + for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) { + struct skb_array *q = band2list(priv, prio); + + bands[prio] = q; + } + + return skb_array_resize_multiple(bands, PFIFO_FAST_BANDS, new_len, + GFP_KERNEL); +} + +struct Qdisc_ops pfifo_fast_ops __read_mostly = { + .id = "pfifo_fast", + .priv_size = sizeof(struct pfifo_fast_priv), + .enqueue = pfifo_fast_enqueue, + .dequeue = pfifo_fast_dequeue, + .peek = pfifo_fast_peek, + .init = pfifo_fast_init, + .destroy = pfifo_fast_destroy, + .reset = pfifo_fast_reset, + .dump = pfifo_fast_dump, + .change_tx_queue_len = pfifo_fast_change_tx_queue_len, + .owner = THIS_MODULE, + .static_flags = TCQ_F_NOLOCK | TCQ_F_CPUSTATS, +}; +EXPORT_SYMBOL(pfifo_fast_ops); + +static struct lock_class_key qdisc_tx_busylock; +static struct lock_class_key qdisc_running_key; + +struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, + const struct Qdisc_ops *ops, + struct netlink_ext_ack *extack) +{ + void *p; + struct Qdisc *sch; + unsigned int size = QDISC_ALIGN(sizeof(*sch)) + ops->priv_size; + int err = -ENOBUFS; + struct net_device *dev; + + if (!dev_queue) { + NL_SET_ERR_MSG(extack, "No device queue given"); + err = -EINVAL; + goto errout; + } + + dev = dev_queue->dev; + p = kzalloc_node(size, GFP_KERNEL, + netdev_queue_numa_node_read(dev_queue)); + + if (!p) + goto errout; + sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p); + /* if we got non aligned memory, ask more and do alignment ourself */ + if (sch != p) { + kfree(p); + p = kzalloc_node(size + QDISC_ALIGNTO - 1, GFP_KERNEL, + netdev_queue_numa_node_read(dev_queue)); + if (!p) + goto errout; + sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p); + sch->padded = (char *) sch - (char *) p; + } + __skb_queue_head_init(&sch->gso_skb); + __skb_queue_head_init(&sch->skb_bad_txq); + qdisc_skb_head_init(&sch->q); + spin_lock_init(&sch->q.lock); + + if (ops->static_flags & TCQ_F_CPUSTATS) { + sch->cpu_bstats = + netdev_alloc_pcpu_stats(struct gnet_stats_basic_cpu); + if (!sch->cpu_bstats) + goto errout1; + + sch->cpu_qstats = alloc_percpu(struct gnet_stats_queue); + if (!sch->cpu_qstats) { + free_percpu(sch->cpu_bstats); + goto errout1; + } + } + + spin_lock_init(&sch->busylock); + lockdep_set_class(&sch->busylock, + dev->qdisc_tx_busylock ?: &qdisc_tx_busylock); + + /* seqlock has the same scope of busylock, for NOLOCK qdisc */ + spin_lock_init(&sch->seqlock); + lockdep_set_class(&sch->busylock, + dev->qdisc_tx_busylock ?: &qdisc_tx_busylock); + + seqcount_init(&sch->running); + lockdep_set_class(&sch->running, + dev->qdisc_running_key ?: &qdisc_running_key); + + sch->ops = ops; + sch->flags = ops->static_flags; + sch->enqueue = ops->enqueue; + sch->dequeue = ops->dequeue; + sch->dev_queue = dev_queue; + dev_hold(dev); + refcount_set(&sch->refcnt, 1); + + return sch; +errout1: + kfree(p); +errout: + return ERR_PTR(err); +} + +struct Qdisc *qdisc_create_dflt(struct netdev_queue *dev_queue, + const struct Qdisc_ops *ops, + unsigned int parentid, + struct netlink_ext_ack *extack) +{ + struct Qdisc *sch; + + if (!try_module_get(ops->owner)) { + NL_SET_ERR_MSG(extack, "Failed to increase module reference counter"); + return NULL; + } + + sch = qdisc_alloc(dev_queue, ops, extack); + if (IS_ERR(sch)) { + module_put(ops->owner); + return NULL; + } + sch->parent = parentid; + + if (!ops->init || ops->init(sch, NULL, extack) == 0) + return sch; + + qdisc_put(sch); + return NULL; +} +EXPORT_SYMBOL(qdisc_create_dflt); + +/* Under qdisc_lock(qdisc) and BH! */ + +void qdisc_reset(struct Qdisc *qdisc) +{ + const struct Qdisc_ops *ops = qdisc->ops; + struct sk_buff *skb, *tmp; + + if (ops->reset) + ops->reset(qdisc); + + skb_queue_walk_safe(&qdisc->gso_skb, skb, tmp) { + __skb_unlink(skb, &qdisc->gso_skb); + kfree_skb_list(skb); + } + + skb_queue_walk_safe(&qdisc->skb_bad_txq, skb, tmp) { + __skb_unlink(skb, &qdisc->skb_bad_txq); + kfree_skb_list(skb); + } + + qdisc->q.qlen = 0; + qdisc->qstats.backlog = 0; +} +EXPORT_SYMBOL(qdisc_reset); + +void qdisc_free(struct Qdisc *qdisc) +{ + if (qdisc_is_percpu_stats(qdisc)) { + free_percpu(qdisc->cpu_bstats); + free_percpu(qdisc->cpu_qstats); + } + + kfree((char *) qdisc - qdisc->padded); +} + +static void qdisc_free_cb(struct rcu_head *head) +{ + struct Qdisc *q = container_of(head, struct Qdisc, rcu); + + qdisc_free(q); +} + +static void qdisc_destroy(struct Qdisc *qdisc) +{ + const struct Qdisc_ops *ops; + struct sk_buff *skb, *tmp; + + if (!qdisc) + return; + ops = qdisc->ops; + +#ifdef CONFIG_NET_SCHED + qdisc_hash_del(qdisc); + + qdisc_put_stab(rtnl_dereference(qdisc->stab)); +#endif + gen_kill_estimator(&qdisc->rate_est); + if (ops->reset) + ops->reset(qdisc); + if (ops->destroy) + ops->destroy(qdisc); + + module_put(ops->owner); + dev_put(qdisc_dev(qdisc)); + + skb_queue_walk_safe(&qdisc->gso_skb, skb, tmp) { + __skb_unlink(skb, &qdisc->gso_skb); + kfree_skb_list(skb); + } + + skb_queue_walk_safe(&qdisc->skb_bad_txq, skb, tmp) { + __skb_unlink(skb, &qdisc->skb_bad_txq); + kfree_skb_list(skb); + } + + call_rcu(&qdisc->rcu, qdisc_free_cb); +} + +void qdisc_put(struct Qdisc *qdisc) +{ + if (qdisc->flags & TCQ_F_BUILTIN || + !refcount_dec_and_test(&qdisc->refcnt)) + return; + + qdisc_destroy(qdisc); +} +EXPORT_SYMBOL(qdisc_put); + +/* Version of qdisc_put() that is called with rtnl mutex unlocked. + * Intended to be used as optimization, this function only takes rtnl lock if + * qdisc reference counter reached zero. + */ + +void qdisc_put_unlocked(struct Qdisc *qdisc) +{ + if (qdisc->flags & TCQ_F_BUILTIN || + !refcount_dec_and_rtnl_lock(&qdisc->refcnt)) + return; + + qdisc_destroy(qdisc); + rtnl_unlock(); +} +EXPORT_SYMBOL(qdisc_put_unlocked); + +/* Attach toplevel qdisc to device queue. */ +struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue, + struct Qdisc *qdisc) +{ + struct Qdisc *oqdisc = dev_queue->qdisc_sleeping; + spinlock_t *root_lock; + + root_lock = qdisc_lock(oqdisc); + spin_lock_bh(root_lock); + + /* ... and graft new one */ + if (qdisc == NULL) + qdisc = &noop_qdisc; + dev_queue->qdisc_sleeping = qdisc; + rcu_assign_pointer(dev_queue->qdisc, &noop_qdisc); + + spin_unlock_bh(root_lock); + + return oqdisc; +} +EXPORT_SYMBOL(dev_graft_qdisc); + +static void attach_one_default_qdisc(struct net_device *dev, + struct netdev_queue *dev_queue, + void *_unused) +{ + struct Qdisc *qdisc; + const struct Qdisc_ops *ops = default_qdisc_ops; + + if (dev->priv_flags & IFF_NO_QUEUE) + ops = &noqueue_qdisc_ops; + + qdisc = qdisc_create_dflt(dev_queue, ops, TC_H_ROOT, NULL); + if (!qdisc) { + netdev_info(dev, "activation failed\n"); + return; + } + if (!netif_is_multiqueue(dev)) + qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + dev_queue->qdisc_sleeping = qdisc; +} + +static void attach_default_qdiscs(struct net_device *dev) +{ + struct netdev_queue *txq; + struct Qdisc *qdisc; + + txq = netdev_get_tx_queue(dev, 0); + + if (!netif_is_multiqueue(dev) || + dev->priv_flags & IFF_NO_QUEUE) { + netdev_for_each_tx_queue(dev, attach_one_default_qdisc, NULL); + dev->qdisc = txq->qdisc_sleeping; + qdisc_refcount_inc(dev->qdisc); + } else { + qdisc = qdisc_create_dflt(txq, &mq_qdisc_ops, TC_H_ROOT, NULL); + if (qdisc) { + dev->qdisc = qdisc; + qdisc->ops->attach(qdisc); + } + } +#ifdef CONFIG_NET_SCHED + if (dev->qdisc != &noop_qdisc) + qdisc_hash_add(dev->qdisc, false); +#endif +} + +static void transition_one_qdisc(struct net_device *dev, + struct netdev_queue *dev_queue, + void *_need_watchdog) +{ + struct Qdisc *new_qdisc = dev_queue->qdisc_sleeping; + int *need_watchdog_p = _need_watchdog; + + if (!(new_qdisc->flags & TCQ_F_BUILTIN)) + clear_bit(__QDISC_STATE_DEACTIVATED, &new_qdisc->state); + + rcu_assign_pointer(dev_queue->qdisc, new_qdisc); + if (need_watchdog_p) { + dev_queue->trans_start = 0; + *need_watchdog_p = 1; + } +} + +void dev_activate(struct net_device *dev) +{ + int need_watchdog; + + /* No queueing discipline is attached to device; + * create default one for devices, which need queueing + * and noqueue_qdisc for virtual interfaces + */ + + if (dev->qdisc == &noop_qdisc) + attach_default_qdiscs(dev); + + if (!netif_carrier_ok(dev)) + /* Delay activation until next carrier-on event */ + return; + + need_watchdog = 0; + netdev_for_each_tx_queue(dev, transition_one_qdisc, &need_watchdog); + if (dev_ingress_queue(dev)) + transition_one_qdisc(dev, dev_ingress_queue(dev), NULL); + + if (need_watchdog) { + netif_trans_update(dev); + dev_watchdog_up(dev); + } +} +EXPORT_SYMBOL(dev_activate); + +static void dev_deactivate_queue(struct net_device *dev, + struct netdev_queue *dev_queue, + void *_qdisc_default) +{ + struct Qdisc *qdisc = rtnl_dereference(dev_queue->qdisc); + struct Qdisc *qdisc_default = _qdisc_default; + + if (qdisc) { + if (!(qdisc->flags & TCQ_F_BUILTIN)) + set_bit(__QDISC_STATE_DEACTIVATED, &qdisc->state); + + rcu_assign_pointer(dev_queue->qdisc, qdisc_default); + } +} + +static void dev_reset_queue(struct net_device *dev, + struct netdev_queue *dev_queue, + void *_unused) +{ + struct Qdisc *qdisc; + bool nolock; + + qdisc = dev_queue->qdisc_sleeping; + if (!qdisc) + return; + + nolock = qdisc->flags & TCQ_F_NOLOCK; + + if (nolock) + spin_lock_bh(&qdisc->seqlock); + spin_lock_bh(qdisc_lock(qdisc)); + + qdisc_reset(qdisc); + + spin_unlock_bh(qdisc_lock(qdisc)); + if (nolock) + spin_unlock_bh(&qdisc->seqlock); +} + +static bool some_qdisc_is_busy(struct net_device *dev) +{ + unsigned int i; + + for (i = 0; i < dev->num_tx_queues; i++) { + struct netdev_queue *dev_queue; + spinlock_t *root_lock; + struct Qdisc *q; + int val; + + dev_queue = netdev_get_tx_queue(dev, i); + q = dev_queue->qdisc_sleeping; + + root_lock = qdisc_lock(q); + spin_lock_bh(root_lock); + + val = (qdisc_is_running(q) || + test_bit(__QDISC_STATE_SCHED, &q->state)); + + spin_unlock_bh(root_lock); + + if (val) + return true; + } + return false; +} + +static void dev_qdisc_reset(struct net_device *dev, + struct netdev_queue *dev_queue, + void *none) +{ + struct Qdisc *qdisc = dev_queue->qdisc_sleeping; + + if (qdisc) + qdisc_reset(qdisc); +} + +/** + * dev_deactivate_many - deactivate transmissions on several devices + * @head: list of devices to deactivate + * + * This function returns only when all outstanding transmissions + * have completed, unless all devices are in dismantle phase. + */ +void dev_deactivate_many(struct list_head *head) +{ + struct net_device *dev; + + list_for_each_entry(dev, head, close_list) { + netdev_for_each_tx_queue(dev, dev_deactivate_queue, + &noop_qdisc); + if (dev_ingress_queue(dev)) + dev_deactivate_queue(dev, dev_ingress_queue(dev), + &noop_qdisc); + + dev_watchdog_down(dev); + } + + /* Wait for outstanding qdisc-less dev_queue_xmit calls or + * outstanding qdisc enqueuing calls. + * This is avoided if all devices are in dismantle phase : + * Caller will call synchronize_net() for us + */ + synchronize_net(); + + list_for_each_entry(dev, head, close_list) { + netdev_for_each_tx_queue(dev, dev_reset_queue, NULL); + + if (dev_ingress_queue(dev)) + dev_reset_queue(dev, dev_ingress_queue(dev), NULL); + } + + /* Wait for outstanding qdisc_run calls. */ + list_for_each_entry(dev, head, close_list) { + while (some_qdisc_is_busy(dev)) + yield(); + /* The new qdisc is assigned at this point so we can safely + * unwind stale skb lists and qdisc statistics + */ + netdev_for_each_tx_queue(dev, dev_qdisc_reset, NULL); + if (dev_ingress_queue(dev)) + dev_qdisc_reset(dev, dev_ingress_queue(dev), NULL); + } +} + +void dev_deactivate(struct net_device *dev) +{ + LIST_HEAD(single); + + list_add(&dev->close_list, &single); + dev_deactivate_many(&single); + list_del(&single); +} +EXPORT_SYMBOL(dev_deactivate); + +static int qdisc_change_tx_queue_len(struct net_device *dev, + struct netdev_queue *dev_queue) +{ + struct Qdisc *qdisc = dev_queue->qdisc_sleeping; + const struct Qdisc_ops *ops = qdisc->ops; + + if (ops->change_tx_queue_len) + return ops->change_tx_queue_len(qdisc, dev->tx_queue_len); + return 0; +} + +void dev_qdisc_change_real_num_tx(struct net_device *dev, + unsigned int new_real_tx) +{ + struct Qdisc *qdisc = dev->qdisc; + + if (qdisc->ops->change_real_num_tx) + qdisc->ops->change_real_num_tx(qdisc, new_real_tx); +} + +int dev_qdisc_change_tx_queue_len(struct net_device *dev) +{ + bool up = dev->flags & IFF_UP; + unsigned int i; + int ret = 0; + + if (up) + dev_deactivate(dev); + + for (i = 0; i < dev->num_tx_queues; i++) { + ret = qdisc_change_tx_queue_len(dev, &dev->_tx[i]); + + /* TODO: revert changes on a partial failure */ + if (ret) + break; + } + + if (up) + dev_activate(dev); + return ret; +} + +static void dev_init_scheduler_queue(struct net_device *dev, + struct netdev_queue *dev_queue, + void *_qdisc) +{ + struct Qdisc *qdisc = _qdisc; + + rcu_assign_pointer(dev_queue->qdisc, qdisc); + dev_queue->qdisc_sleeping = qdisc; +} + +void dev_init_scheduler(struct net_device *dev) +{ + dev->qdisc = &noop_qdisc; + netdev_for_each_tx_queue(dev, dev_init_scheduler_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + dev_init_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); + + timer_setup(&dev->watchdog_timer, dev_watchdog, 0); +} + +static void shutdown_scheduler_queue(struct net_device *dev, + struct netdev_queue *dev_queue, + void *_qdisc_default) +{ + struct Qdisc *qdisc = dev_queue->qdisc_sleeping; + struct Qdisc *qdisc_default = _qdisc_default; + + if (qdisc) { + rcu_assign_pointer(dev_queue->qdisc, qdisc_default); + dev_queue->qdisc_sleeping = qdisc_default; + + qdisc_put(qdisc); + } +} + +void dev_shutdown(struct net_device *dev) +{ + netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc); + if (dev_ingress_queue(dev)) + shutdown_scheduler_queue(dev, dev_ingress_queue(dev), &noop_qdisc); + qdisc_put(dev->qdisc); + dev->qdisc = &noop_qdisc; + + WARN_ON(timer_pending(&dev->watchdog_timer)); +} + +void psched_ratecfg_precompute(struct psched_ratecfg *r, + const struct tc_ratespec *conf, + u64 rate64) +{ + memset(r, 0, sizeof(*r)); + r->overhead = conf->overhead; + r->mpu = conf->mpu; + r->rate_bytes_ps = max_t(u64, conf->rate, rate64); + r->linklayer = (conf->linklayer & TC_LINKLAYER_MASK); + r->mult = 1; + /* + * The deal here is to replace a divide by a reciprocal one + * in fast path (a reciprocal divide is a multiply and a shift) + * + * Normal formula would be : + * time_in_ns = (NSEC_PER_SEC * len) / rate_bps + * + * We compute mult/shift to use instead : + * time_in_ns = (len * mult) >> shift; + * + * We try to get the highest possible mult value for accuracy, + * but have to make sure no overflows will ever happen. + */ + if (r->rate_bytes_ps > 0) { + u64 factor = NSEC_PER_SEC; + + for (;;) { + r->mult = div64_u64(factor, r->rate_bytes_ps); + if (r->mult & (1U << 31) || factor & (1ULL << 63)) + break; + factor <<= 1; + r->shift++; + } + } +} +EXPORT_SYMBOL(psched_ratecfg_precompute); + +static void mini_qdisc_rcu_func(struct rcu_head *head) +{ +} + +void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp, + struct tcf_proto *tp_head) +{ + struct mini_Qdisc *miniq_old = rtnl_dereference(*miniqp->p_miniq); + struct mini_Qdisc *miniq; + + if (!tp_head) { + RCU_INIT_POINTER(*miniqp->p_miniq, NULL); + /* Wait for flying RCU callback before it is freed. */ + rcu_barrier_bh(); + return; + } + + miniq = !miniq_old || miniq_old == &miniqp->miniq2 ? + &miniqp->miniq1 : &miniqp->miniq2; + + /* We need to make sure that readers won't see the miniq + * we are about to modify. So wait until previous call_rcu_bh callback + * is done. + */ + rcu_barrier_bh(); + miniq->filter_list = tp_head; + rcu_assign_pointer(*miniqp->p_miniq, miniq); + + if (miniq_old) + /* This is counterpart of the rcu barriers above. We need to + * block potential new user of miniq_old until all readers + * are not seeing it. + */ + call_rcu_bh(&miniq_old->rcu, mini_qdisc_rcu_func); +} +EXPORT_SYMBOL(mini_qdisc_pair_swap); + +void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc, + struct mini_Qdisc __rcu **p_miniq) +{ + miniqp->miniq1.cpu_bstats = qdisc->cpu_bstats; + miniqp->miniq1.cpu_qstats = qdisc->cpu_qstats; + miniqp->miniq2.cpu_bstats = qdisc->cpu_bstats; + miniqp->miniq2.cpu_qstats = qdisc->cpu_qstats; + miniqp->p_miniq = p_miniq; +} +EXPORT_SYMBOL(mini_qdisc_pair_init); diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c new file mode 100644 index 000000000..db7a91a29 --- /dev/null +++ b/net/sched/sch_gred.c @@ -0,0 +1,620 @@ +/* + * net/sched/sch_gred.c Generic Random Early Detection queue. + * + * + * 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. + * + * Authors: J Hadi Salim (hadi@cyberus.ca) 1998-2002 + * + * 991129: - Bug fix with grio mode + * - a better sing. AvgQ mode with Grio(WRED) + * - A finer grained VQ dequeue based on sugestion + * from Ren Liu + * - More error checks + * + * For all the glorious comments look at include/net/red.h + */ + +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> +#include <net/red.h> + +#define GRED_DEF_PRIO (MAX_DPs / 2) +#define GRED_VQ_MASK (MAX_DPs - 1) + +struct gred_sched_data; +struct gred_sched; + +struct gred_sched_data { + u32 limit; /* HARD maximal queue length */ + u32 DP; /* the drop parameters */ + u32 bytesin; /* bytes seen on virtualQ so far*/ + u32 packetsin; /* packets seen on virtualQ so far*/ + u32 backlog; /* bytes on the virtualQ */ + u8 prio; /* the prio of this vq */ + + struct red_parms parms; + struct red_vars vars; + struct red_stats stats; +}; + +enum { + GRED_WRED_MODE = 1, + GRED_RIO_MODE, +}; + +struct gred_sched { + struct gred_sched_data *tab[MAX_DPs]; + unsigned long flags; + u32 red_flags; + u32 DPs; + u32 def; + struct red_vars wred_set; +}; + +static inline int gred_wred_mode(struct gred_sched *table) +{ + return test_bit(GRED_WRED_MODE, &table->flags); +} + +static inline void gred_enable_wred_mode(struct gred_sched *table) +{ + __set_bit(GRED_WRED_MODE, &table->flags); +} + +static inline void gred_disable_wred_mode(struct gred_sched *table) +{ + __clear_bit(GRED_WRED_MODE, &table->flags); +} + +static inline int gred_rio_mode(struct gred_sched *table) +{ + return test_bit(GRED_RIO_MODE, &table->flags); +} + +static inline void gred_enable_rio_mode(struct gred_sched *table) +{ + __set_bit(GRED_RIO_MODE, &table->flags); +} + +static inline void gred_disable_rio_mode(struct gred_sched *table) +{ + __clear_bit(GRED_RIO_MODE, &table->flags); +} + +static inline int gred_wred_mode_check(struct Qdisc *sch) +{ + struct gred_sched *table = qdisc_priv(sch); + int i; + + /* Really ugly O(n^2) but shouldn't be necessary too frequent. */ + for (i = 0; i < table->DPs; i++) { + struct gred_sched_data *q = table->tab[i]; + int n; + + if (q == NULL) + continue; + + for (n = i + 1; n < table->DPs; n++) + if (table->tab[n] && table->tab[n]->prio == q->prio) + return 1; + } + + return 0; +} + +static inline unsigned int gred_backlog(struct gred_sched *table, + struct gred_sched_data *q, + struct Qdisc *sch) +{ + if (gred_wred_mode(table)) + return sch->qstats.backlog; + else + return q->backlog; +} + +static inline u16 tc_index_to_dp(struct sk_buff *skb) +{ + return skb->tc_index & GRED_VQ_MASK; +} + +static inline void gred_load_wred_set(const struct gred_sched *table, + struct gred_sched_data *q) +{ + q->vars.qavg = table->wred_set.qavg; + q->vars.qidlestart = table->wred_set.qidlestart; +} + +static inline void gred_store_wred_set(struct gred_sched *table, + struct gred_sched_data *q) +{ + table->wred_set.qavg = q->vars.qavg; + table->wred_set.qidlestart = q->vars.qidlestart; +} + +static inline int gred_use_ecn(struct gred_sched *t) +{ + return t->red_flags & TC_RED_ECN; +} + +static inline int gred_use_harddrop(struct gred_sched *t) +{ + return t->red_flags & TC_RED_HARDDROP; +} + +static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct gred_sched_data *q = NULL; + struct gred_sched *t = qdisc_priv(sch); + unsigned long qavg = 0; + u16 dp = tc_index_to_dp(skb); + + if (dp >= t->DPs || (q = t->tab[dp]) == NULL) { + dp = t->def; + + q = t->tab[dp]; + if (!q) { + /* Pass through packets not assigned to a DP + * if no default DP has been configured. This + * allows for DP flows to be left untouched. + */ + if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= + sch->limit)) + return qdisc_enqueue_tail(skb, sch); + else + goto drop; + } + + /* fix tc_index? --could be controversial but needed for + requeueing */ + skb->tc_index = (skb->tc_index & ~GRED_VQ_MASK) | dp; + } + + /* sum up all the qaves of prios < ours to get the new qave */ + if (!gred_wred_mode(t) && gred_rio_mode(t)) { + int i; + + for (i = 0; i < t->DPs; i++) { + if (t->tab[i] && t->tab[i]->prio < q->prio && + !red_is_idling(&t->tab[i]->vars)) + qavg += t->tab[i]->vars.qavg; + } + + } + + q->packetsin++; + q->bytesin += qdisc_pkt_len(skb); + + if (gred_wred_mode(t)) + gred_load_wred_set(t, q); + + q->vars.qavg = red_calc_qavg(&q->parms, + &q->vars, + gred_backlog(t, q, sch)); + + if (red_is_idling(&q->vars)) + red_end_of_idle_period(&q->vars); + + if (gred_wred_mode(t)) + gred_store_wred_set(t, q); + + switch (red_action(&q->parms, &q->vars, q->vars.qavg + qavg)) { + case RED_DONT_MARK: + break; + + case RED_PROB_MARK: + qdisc_qstats_overlimit(sch); + if (!gred_use_ecn(t) || !INET_ECN_set_ce(skb)) { + q->stats.prob_drop++; + goto congestion_drop; + } + + q->stats.prob_mark++; + break; + + case RED_HARD_MARK: + qdisc_qstats_overlimit(sch); + if (gred_use_harddrop(t) || !gred_use_ecn(t) || + !INET_ECN_set_ce(skb)) { + q->stats.forced_drop++; + goto congestion_drop; + } + q->stats.forced_mark++; + break; + } + + if (gred_backlog(t, q, sch) + qdisc_pkt_len(skb) <= q->limit) { + q->backlog += qdisc_pkt_len(skb); + return qdisc_enqueue_tail(skb, sch); + } + + q->stats.pdrop++; +drop: + return qdisc_drop(skb, sch, to_free); + +congestion_drop: + qdisc_drop(skb, sch, to_free); + return NET_XMIT_CN; +} + +static struct sk_buff *gred_dequeue(struct Qdisc *sch) +{ + struct sk_buff *skb; + struct gred_sched *t = qdisc_priv(sch); + + skb = qdisc_dequeue_head(sch); + + if (skb) { + struct gred_sched_data *q; + u16 dp = tc_index_to_dp(skb); + + if (dp >= t->DPs || (q = t->tab[dp]) == NULL) { + net_warn_ratelimited("GRED: Unable to relocate VQ 0x%x after dequeue, screwing up backlog\n", + tc_index_to_dp(skb)); + } else { + q->backlog -= qdisc_pkt_len(skb); + + if (gred_wred_mode(t)) { + if (!sch->qstats.backlog) + red_start_of_idle_period(&t->wred_set); + } else { + if (!q->backlog) + red_start_of_idle_period(&q->vars); + } + } + + return skb; + } + + return NULL; +} + +static void gred_reset(struct Qdisc *sch) +{ + int i; + struct gred_sched *t = qdisc_priv(sch); + + qdisc_reset_queue(sch); + + for (i = 0; i < t->DPs; i++) { + struct gred_sched_data *q = t->tab[i]; + + if (!q) + continue; + + red_restart(&q->vars); + q->backlog = 0; + } +} + +static inline void gred_destroy_vq(struct gred_sched_data *q) +{ + kfree(q); +} + +static inline int gred_change_table_def(struct Qdisc *sch, struct nlattr *dps) +{ + struct gred_sched *table = qdisc_priv(sch); + struct tc_gred_sopt *sopt; + int i; + + if (!dps) + return -EINVAL; + + sopt = nla_data(dps); + + if (sopt->DPs > MAX_DPs || sopt->DPs == 0 || + sopt->def_DP >= sopt->DPs) + return -EINVAL; + + sch_tree_lock(sch); + table->DPs = sopt->DPs; + table->def = sopt->def_DP; + table->red_flags = sopt->flags; + + /* + * Every entry point to GRED is synchronized with the above code + * and the DP is checked against DPs, i.e. shadowed VQs can no + * longer be found so we can unlock right here. + */ + sch_tree_unlock(sch); + + if (sopt->grio) { + gred_enable_rio_mode(table); + gred_disable_wred_mode(table); + if (gred_wred_mode_check(sch)) + gred_enable_wred_mode(table); + } else { + gred_disable_rio_mode(table); + gred_disable_wred_mode(table); + } + + for (i = table->DPs; i < MAX_DPs; i++) { + if (table->tab[i]) { + pr_warn("GRED: Warning: Destroying shadowed VQ 0x%x\n", + i); + gred_destroy_vq(table->tab[i]); + table->tab[i] = NULL; + } + } + + return 0; +} + +static inline int gred_change_vq(struct Qdisc *sch, int dp, + struct tc_gred_qopt *ctl, int prio, + u8 *stab, u32 max_P, + struct gred_sched_data **prealloc) +{ + struct gred_sched *table = qdisc_priv(sch); + struct gred_sched_data *q = table->tab[dp]; + + if (!red_check_params(ctl->qth_min, ctl->qth_max, ctl->Wlog, ctl->Scell_log, stab)) + return -EINVAL; + + if (!q) { + table->tab[dp] = q = *prealloc; + *prealloc = NULL; + if (!q) + return -ENOMEM; + } + + q->DP = dp; + q->prio = prio; + if (ctl->limit > sch->limit) + q->limit = sch->limit; + else + q->limit = ctl->limit; + + if (q->backlog == 0) + red_end_of_idle_period(&q->vars); + + red_set_parms(&q->parms, + ctl->qth_min, ctl->qth_max, ctl->Wlog, ctl->Plog, + ctl->Scell_log, stab, max_P); + red_set_vars(&q->vars); + return 0; +} + +static const struct nla_policy gred_policy[TCA_GRED_MAX + 1] = { + [TCA_GRED_PARMS] = { .len = sizeof(struct tc_gred_qopt) }, + [TCA_GRED_STAB] = { .len = 256 }, + [TCA_GRED_DPS] = { .len = sizeof(struct tc_gred_sopt) }, + [TCA_GRED_MAX_P] = { .type = NLA_U32 }, + [TCA_GRED_LIMIT] = { .type = NLA_U32 }, +}; + +static int gred_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct gred_sched *table = qdisc_priv(sch); + struct tc_gred_qopt *ctl; + struct nlattr *tb[TCA_GRED_MAX + 1]; + int err, prio = GRED_DEF_PRIO; + u8 *stab; + u32 max_P; + struct gred_sched_data *prealloc; + + if (opt == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_GRED_MAX, opt, gred_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_GRED_PARMS] == NULL && tb[TCA_GRED_STAB] == NULL) { + if (tb[TCA_GRED_LIMIT] != NULL) + sch->limit = nla_get_u32(tb[TCA_GRED_LIMIT]); + return gred_change_table_def(sch, tb[TCA_GRED_DPS]); + } + + if (tb[TCA_GRED_PARMS] == NULL || + tb[TCA_GRED_STAB] == NULL || + tb[TCA_GRED_LIMIT] != NULL) + return -EINVAL; + + max_P = tb[TCA_GRED_MAX_P] ? nla_get_u32(tb[TCA_GRED_MAX_P]) : 0; + + err = -EINVAL; + ctl = nla_data(tb[TCA_GRED_PARMS]); + stab = nla_data(tb[TCA_GRED_STAB]); + + if (ctl->DP >= table->DPs) + goto errout; + + if (gred_rio_mode(table)) { + if (ctl->prio == 0) { + int def_prio = GRED_DEF_PRIO; + + if (table->tab[table->def]) + def_prio = table->tab[table->def]->prio; + + printk(KERN_DEBUG "GRED: DP %u does not have a prio " + "setting default to %d\n", ctl->DP, def_prio); + + prio = def_prio; + } else + prio = ctl->prio; + } + + prealloc = kzalloc(sizeof(*prealloc), GFP_KERNEL); + sch_tree_lock(sch); + + err = gred_change_vq(sch, ctl->DP, ctl, prio, stab, max_P, &prealloc); + if (err < 0) + goto errout_locked; + + if (gred_rio_mode(table)) { + gred_disable_wred_mode(table); + if (gred_wred_mode_check(sch)) + gred_enable_wred_mode(table); + } + + err = 0; + +errout_locked: + sch_tree_unlock(sch); + kfree(prealloc); +errout: + return err; +} + +static int gred_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_GRED_MAX + 1]; + int err; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_GRED_MAX, opt, gred_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_GRED_PARMS] || tb[TCA_GRED_STAB]) + return -EINVAL; + + if (tb[TCA_GRED_LIMIT]) + sch->limit = nla_get_u32(tb[TCA_GRED_LIMIT]); + else + sch->limit = qdisc_dev(sch)->tx_queue_len + * psched_mtu(qdisc_dev(sch)); + + return gred_change_table_def(sch, tb[TCA_GRED_DPS]); +} + +static int gred_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct gred_sched *table = qdisc_priv(sch); + struct nlattr *parms, *opts = NULL; + int i; + u32 max_p[MAX_DPs]; + struct tc_gred_sopt sopt = { + .DPs = table->DPs, + .def_DP = table->def, + .grio = gred_rio_mode(table), + .flags = table->red_flags, + }; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + if (nla_put(skb, TCA_GRED_DPS, sizeof(sopt), &sopt)) + goto nla_put_failure; + + for (i = 0; i < MAX_DPs; i++) { + struct gred_sched_data *q = table->tab[i]; + + max_p[i] = q ? q->parms.max_P : 0; + } + if (nla_put(skb, TCA_GRED_MAX_P, sizeof(max_p), max_p)) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_GRED_LIMIT, sch->limit)) + goto nla_put_failure; + + parms = nla_nest_start(skb, TCA_GRED_PARMS); + if (parms == NULL) + goto nla_put_failure; + + for (i = 0; i < MAX_DPs; i++) { + struct gred_sched_data *q = table->tab[i]; + struct tc_gred_qopt opt; + unsigned long qavg; + + memset(&opt, 0, sizeof(opt)); + + if (!q) { + /* hack -- fix at some point with proper message + This is how we indicate to tc that there is no VQ + at this DP */ + + opt.DP = MAX_DPs + i; + goto append_opt; + } + + opt.limit = q->limit; + opt.DP = q->DP; + opt.backlog = gred_backlog(table, q, sch); + opt.prio = q->prio; + opt.qth_min = q->parms.qth_min >> q->parms.Wlog; + opt.qth_max = q->parms.qth_max >> q->parms.Wlog; + opt.Wlog = q->parms.Wlog; + opt.Plog = q->parms.Plog; + opt.Scell_log = q->parms.Scell_log; + opt.other = q->stats.other; + opt.early = q->stats.prob_drop; + opt.forced = q->stats.forced_drop; + opt.pdrop = q->stats.pdrop; + opt.packets = q->packetsin; + opt.bytesin = q->bytesin; + + if (gred_wred_mode(table)) + gred_load_wred_set(table, q); + + qavg = red_calc_qavg(&q->parms, &q->vars, + q->vars.qavg >> q->parms.Wlog); + opt.qave = qavg >> q->parms.Wlog; + +append_opt: + if (nla_append(skb, sizeof(opt), &opt) < 0) + goto nla_put_failure; + } + + nla_nest_end(skb, parms); + + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -EMSGSIZE; +} + +static void gred_destroy(struct Qdisc *sch) +{ + struct gred_sched *table = qdisc_priv(sch); + int i; + + for (i = 0; i < table->DPs; i++) { + if (table->tab[i]) + gred_destroy_vq(table->tab[i]); + } +} + +static struct Qdisc_ops gred_qdisc_ops __read_mostly = { + .id = "gred", + .priv_size = sizeof(struct gred_sched), + .enqueue = gred_enqueue, + .dequeue = gred_dequeue, + .peek = qdisc_peek_head, + .init = gred_init, + .reset = gred_reset, + .destroy = gred_destroy, + .change = gred_change, + .dump = gred_dump, + .owner = THIS_MODULE, +}; + +static int __init gred_module_init(void) +{ + return register_qdisc(&gred_qdisc_ops); +} + +static void __exit gred_module_exit(void) +{ + unregister_qdisc(&gred_qdisc_ops); +} + +module_init(gred_module_init) +module_exit(gred_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c new file mode 100644 index 000000000..b18ec1f6d --- /dev/null +++ b/net/sched/sch_hfsc.c @@ -0,0 +1,1697 @@ +/* + * Copyright (c) 2003 Patrick McHardy, <kaber@trash.net> + * + * 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. + * + * 2003-10-17 - Ported from altq + */ +/* + * Copyright (c) 1997-1999 Carnegie Mellon University. All Rights Reserved. + * + * Permission to use, copy, modify, and distribute this software and + * its documentation is hereby granted (including for commercial or + * for-profit use), provided that both the copyright notice and this + * permission notice appear in all copies of the software, derivative + * works, or modified versions, and any portions thereof. + * + * THIS SOFTWARE IS EXPERIMENTAL AND IS KNOWN TO HAVE BUGS, SOME OF + * WHICH MAY HAVE SERIOUS CONSEQUENCES. CARNEGIE MELLON PROVIDES THIS + * SOFTWARE IN ITS ``AS IS'' CONDITION, AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL CARNEGIE MELLON UNIVERSITY BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT + * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE + * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH + * DAMAGE. + * + * Carnegie Mellon encourages (but does not require) users of this + * software to return any improvements or extensions that they make, + * and to grant Carnegie Mellon the rights to redistribute these + * changes without encumbrance. + */ +/* + * H-FSC is described in Proceedings of SIGCOMM'97, + * "A Hierarchical Fair Service Curve Algorithm for Link-Sharing, + * Real-Time and Priority Service" + * by Ion Stoica, Hui Zhang, and T. S. Eugene Ng. + * + * Oleg Cherevko <olwi@aq.ml.com.ua> added the upperlimit for link-sharing. + * when a class has an upperlimit, the fit-time is computed from the + * upperlimit service curve. the link-sharing scheduler does not schedule + * a class whose fit-time exceeds the current time. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/compiler.h> +#include <linux/spinlock.h> +#include <linux/skbuff.h> +#include <linux/string.h> +#include <linux/slab.h> +#include <linux/list.h> +#include <linux/rbtree.h> +#include <linux/init.h> +#include <linux/rtnetlink.h> +#include <linux/pkt_sched.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <asm/div64.h> + +/* + * kernel internal service curve representation: + * coordinates are given by 64 bit unsigned integers. + * x-axis: unit is clock count. + * y-axis: unit is byte. + * + * The service curve parameters are converted to the internal + * representation. The slope values are scaled to avoid overflow. + * the inverse slope values as well as the y-projection of the 1st + * segment are kept in order to avoid 64-bit divide operations + * that are expensive on 32-bit architectures. + */ + +struct internal_sc { + u64 sm1; /* scaled slope of the 1st segment */ + u64 ism1; /* scaled inverse-slope of the 1st segment */ + u64 dx; /* the x-projection of the 1st segment */ + u64 dy; /* the y-projection of the 1st segment */ + u64 sm2; /* scaled slope of the 2nd segment */ + u64 ism2; /* scaled inverse-slope of the 2nd segment */ +}; + +/* runtime service curve */ +struct runtime_sc { + u64 x; /* current starting position on x-axis */ + u64 y; /* current starting position on y-axis */ + u64 sm1; /* scaled slope of the 1st segment */ + u64 ism1; /* scaled inverse-slope of the 1st segment */ + u64 dx; /* the x-projection of the 1st segment */ + u64 dy; /* the y-projection of the 1st segment */ + u64 sm2; /* scaled slope of the 2nd segment */ + u64 ism2; /* scaled inverse-slope of the 2nd segment */ +}; + +enum hfsc_class_flags { + HFSC_RSC = 0x1, + HFSC_FSC = 0x2, + HFSC_USC = 0x4 +}; + +struct hfsc_class { + struct Qdisc_class_common cl_common; + + struct gnet_stats_basic_packed bstats; + struct gnet_stats_queue qstats; + struct net_rate_estimator __rcu *rate_est; + struct tcf_proto __rcu *filter_list; /* filter list */ + struct tcf_block *block; + unsigned int filter_cnt; /* filter count */ + unsigned int level; /* class level in hierarchy */ + + struct hfsc_sched *sched; /* scheduler data */ + struct hfsc_class *cl_parent; /* parent class */ + struct list_head siblings; /* sibling classes */ + struct list_head children; /* child classes */ + struct Qdisc *qdisc; /* leaf qdisc */ + + struct rb_node el_node; /* qdisc's eligible tree member */ + struct rb_root vt_tree; /* active children sorted by cl_vt */ + struct rb_node vt_node; /* parent's vt_tree member */ + struct rb_root cf_tree; /* active children sorted by cl_f */ + struct rb_node cf_node; /* parent's cf_heap member */ + + u64 cl_total; /* total work in bytes */ + u64 cl_cumul; /* cumulative work in bytes done by + real-time criteria */ + + u64 cl_d; /* deadline*/ + u64 cl_e; /* eligible time */ + u64 cl_vt; /* virtual time */ + u64 cl_f; /* time when this class will fit for + link-sharing, max(myf, cfmin) */ + u64 cl_myf; /* my fit-time (calculated from this + class's own upperlimit curve) */ + u64 cl_cfmin; /* earliest children's fit-time (used + with cl_myf to obtain cl_f) */ + u64 cl_cvtmin; /* minimal virtual time among the + children fit for link-sharing + (monotonic within a period) */ + u64 cl_vtadj; /* intra-period cumulative vt + adjustment */ + u64 cl_cvtoff; /* largest virtual time seen among + the children */ + + struct internal_sc cl_rsc; /* internal real-time service curve */ + struct internal_sc cl_fsc; /* internal fair service curve */ + struct internal_sc cl_usc; /* internal upperlimit service curve */ + struct runtime_sc cl_deadline; /* deadline curve */ + struct runtime_sc cl_eligible; /* eligible curve */ + struct runtime_sc cl_virtual; /* virtual curve */ + struct runtime_sc cl_ulimit; /* upperlimit curve */ + + u8 cl_flags; /* which curves are valid */ + u32 cl_vtperiod; /* vt period sequence number */ + u32 cl_parentperiod;/* parent's vt period sequence number*/ + u32 cl_nactive; /* number of active children */ +}; + +struct hfsc_sched { + u16 defcls; /* default class id */ + struct hfsc_class root; /* root class */ + struct Qdisc_class_hash clhash; /* class hash */ + struct rb_root eligible; /* eligible tree */ + struct qdisc_watchdog watchdog; /* watchdog timer */ +}; + +#define HT_INFINITY 0xffffffffffffffffULL /* infinite time value */ + + +/* + * eligible tree holds backlogged classes being sorted by their eligible times. + * there is one eligible tree per hfsc instance. + */ + +static void +eltree_insert(struct hfsc_class *cl) +{ + struct rb_node **p = &cl->sched->eligible.rb_node; + struct rb_node *parent = NULL; + struct hfsc_class *cl1; + + while (*p != NULL) { + parent = *p; + cl1 = rb_entry(parent, struct hfsc_class, el_node); + if (cl->cl_e >= cl1->cl_e) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&cl->el_node, parent, p); + rb_insert_color(&cl->el_node, &cl->sched->eligible); +} + +static inline void +eltree_remove(struct hfsc_class *cl) +{ + rb_erase(&cl->el_node, &cl->sched->eligible); +} + +static inline void +eltree_update(struct hfsc_class *cl) +{ + eltree_remove(cl); + eltree_insert(cl); +} + +/* find the class with the minimum deadline among the eligible classes */ +static inline struct hfsc_class * +eltree_get_mindl(struct hfsc_sched *q, u64 cur_time) +{ + struct hfsc_class *p, *cl = NULL; + struct rb_node *n; + + for (n = rb_first(&q->eligible); n != NULL; n = rb_next(n)) { + p = rb_entry(n, struct hfsc_class, el_node); + if (p->cl_e > cur_time) + break; + if (cl == NULL || p->cl_d < cl->cl_d) + cl = p; + } + return cl; +} + +/* find the class with minimum eligible time among the eligible classes */ +static inline struct hfsc_class * +eltree_get_minel(struct hfsc_sched *q) +{ + struct rb_node *n; + + n = rb_first(&q->eligible); + if (n == NULL) + return NULL; + return rb_entry(n, struct hfsc_class, el_node); +} + +/* + * vttree holds holds backlogged child classes being sorted by their virtual + * time. each intermediate class has one vttree. + */ +static void +vttree_insert(struct hfsc_class *cl) +{ + struct rb_node **p = &cl->cl_parent->vt_tree.rb_node; + struct rb_node *parent = NULL; + struct hfsc_class *cl1; + + while (*p != NULL) { + parent = *p; + cl1 = rb_entry(parent, struct hfsc_class, vt_node); + if (cl->cl_vt >= cl1->cl_vt) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&cl->vt_node, parent, p); + rb_insert_color(&cl->vt_node, &cl->cl_parent->vt_tree); +} + +static inline void +vttree_remove(struct hfsc_class *cl) +{ + rb_erase(&cl->vt_node, &cl->cl_parent->vt_tree); +} + +static inline void +vttree_update(struct hfsc_class *cl) +{ + vttree_remove(cl); + vttree_insert(cl); +} + +static inline struct hfsc_class * +vttree_firstfit(struct hfsc_class *cl, u64 cur_time) +{ + struct hfsc_class *p; + struct rb_node *n; + + for (n = rb_first(&cl->vt_tree); n != NULL; n = rb_next(n)) { + p = rb_entry(n, struct hfsc_class, vt_node); + if (p->cl_f <= cur_time) + return p; + } + return NULL; +} + +/* + * get the leaf class with the minimum vt in the hierarchy + */ +static struct hfsc_class * +vttree_get_minvt(struct hfsc_class *cl, u64 cur_time) +{ + /* if root-class's cfmin is bigger than cur_time nothing to do */ + if (cl->cl_cfmin > cur_time) + return NULL; + + while (cl->level > 0) { + cl = vttree_firstfit(cl, cur_time); + if (cl == NULL) + return NULL; + /* + * update parent's cl_cvtmin. + */ + if (cl->cl_parent->cl_cvtmin < cl->cl_vt) + cl->cl_parent->cl_cvtmin = cl->cl_vt; + } + return cl; +} + +static void +cftree_insert(struct hfsc_class *cl) +{ + struct rb_node **p = &cl->cl_parent->cf_tree.rb_node; + struct rb_node *parent = NULL; + struct hfsc_class *cl1; + + while (*p != NULL) { + parent = *p; + cl1 = rb_entry(parent, struct hfsc_class, cf_node); + if (cl->cl_f >= cl1->cl_f) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&cl->cf_node, parent, p); + rb_insert_color(&cl->cf_node, &cl->cl_parent->cf_tree); +} + +static inline void +cftree_remove(struct hfsc_class *cl) +{ + rb_erase(&cl->cf_node, &cl->cl_parent->cf_tree); +} + +static inline void +cftree_update(struct hfsc_class *cl) +{ + cftree_remove(cl); + cftree_insert(cl); +} + +/* + * service curve support functions + * + * external service curve parameters + * m: bps + * d: us + * internal service curve parameters + * sm: (bytes/psched_us) << SM_SHIFT + * ism: (psched_us/byte) << ISM_SHIFT + * dx: psched_us + * + * The clock source resolution with ktime and PSCHED_SHIFT 10 is 1.024us. + * + * sm and ism are scaled in order to keep effective digits. + * SM_SHIFT and ISM_SHIFT are selected to keep at least 4 effective + * digits in decimal using the following table. + * + * bits/sec 100Kbps 1Mbps 10Mbps 100Mbps 1Gbps + * ------------+------------------------------------------------------- + * bytes/1.024us 12.8e-3 128e-3 1280e-3 12800e-3 128000e-3 + * + * 1.024us/byte 78.125 7.8125 0.78125 0.078125 0.0078125 + * + * So, for PSCHED_SHIFT 10 we need: SM_SHIFT 20, ISM_SHIFT 18. + */ +#define SM_SHIFT (30 - PSCHED_SHIFT) +#define ISM_SHIFT (8 + PSCHED_SHIFT) + +#define SM_MASK ((1ULL << SM_SHIFT) - 1) +#define ISM_MASK ((1ULL << ISM_SHIFT) - 1) + +static inline u64 +seg_x2y(u64 x, u64 sm) +{ + u64 y; + + /* + * compute + * y = x * sm >> SM_SHIFT + * but divide it for the upper and lower bits to avoid overflow + */ + y = (x >> SM_SHIFT) * sm + (((x & SM_MASK) * sm) >> SM_SHIFT); + return y; +} + +static inline u64 +seg_y2x(u64 y, u64 ism) +{ + u64 x; + + if (y == 0) + x = 0; + else if (ism == HT_INFINITY) + x = HT_INFINITY; + else { + x = (y >> ISM_SHIFT) * ism + + (((y & ISM_MASK) * ism) >> ISM_SHIFT); + } + return x; +} + +/* Convert m (bps) into sm (bytes/psched us) */ +static u64 +m2sm(u32 m) +{ + u64 sm; + + sm = ((u64)m << SM_SHIFT); + sm += PSCHED_TICKS_PER_SEC - 1; + do_div(sm, PSCHED_TICKS_PER_SEC); + return sm; +} + +/* convert m (bps) into ism (psched us/byte) */ +static u64 +m2ism(u32 m) +{ + u64 ism; + + if (m == 0) + ism = HT_INFINITY; + else { + ism = ((u64)PSCHED_TICKS_PER_SEC << ISM_SHIFT); + ism += m - 1; + do_div(ism, m); + } + return ism; +} + +/* convert d (us) into dx (psched us) */ +static u64 +d2dx(u32 d) +{ + u64 dx; + + dx = ((u64)d * PSCHED_TICKS_PER_SEC); + dx += USEC_PER_SEC - 1; + do_div(dx, USEC_PER_SEC); + return dx; +} + +/* convert sm (bytes/psched us) into m (bps) */ +static u32 +sm2m(u64 sm) +{ + u64 m; + + m = (sm * PSCHED_TICKS_PER_SEC) >> SM_SHIFT; + return (u32)m; +} + +/* convert dx (psched us) into d (us) */ +static u32 +dx2d(u64 dx) +{ + u64 d; + + d = dx * USEC_PER_SEC; + do_div(d, PSCHED_TICKS_PER_SEC); + return (u32)d; +} + +static void +sc2isc(struct tc_service_curve *sc, struct internal_sc *isc) +{ + isc->sm1 = m2sm(sc->m1); + isc->ism1 = m2ism(sc->m1); + isc->dx = d2dx(sc->d); + isc->dy = seg_x2y(isc->dx, isc->sm1); + isc->sm2 = m2sm(sc->m2); + isc->ism2 = m2ism(sc->m2); +} + +/* + * initialize the runtime service curve with the given internal + * service curve starting at (x, y). + */ +static void +rtsc_init(struct runtime_sc *rtsc, struct internal_sc *isc, u64 x, u64 y) +{ + rtsc->x = x; + rtsc->y = y; + rtsc->sm1 = isc->sm1; + rtsc->ism1 = isc->ism1; + rtsc->dx = isc->dx; + rtsc->dy = isc->dy; + rtsc->sm2 = isc->sm2; + rtsc->ism2 = isc->ism2; +} + +/* + * calculate the y-projection of the runtime service curve by the + * given x-projection value + */ +static u64 +rtsc_y2x(struct runtime_sc *rtsc, u64 y) +{ + u64 x; + + if (y < rtsc->y) + x = rtsc->x; + else if (y <= rtsc->y + rtsc->dy) { + /* x belongs to the 1st segment */ + if (rtsc->dy == 0) + x = rtsc->x + rtsc->dx; + else + x = rtsc->x + seg_y2x(y - rtsc->y, rtsc->ism1); + } else { + /* x belongs to the 2nd segment */ + x = rtsc->x + rtsc->dx + + seg_y2x(y - rtsc->y - rtsc->dy, rtsc->ism2); + } + return x; +} + +static u64 +rtsc_x2y(struct runtime_sc *rtsc, u64 x) +{ + u64 y; + + if (x <= rtsc->x) + y = rtsc->y; + else if (x <= rtsc->x + rtsc->dx) + /* y belongs to the 1st segment */ + y = rtsc->y + seg_x2y(x - rtsc->x, rtsc->sm1); + else + /* y belongs to the 2nd segment */ + y = rtsc->y + rtsc->dy + + seg_x2y(x - rtsc->x - rtsc->dx, rtsc->sm2); + return y; +} + +/* + * update the runtime service curve by taking the minimum of the current + * runtime service curve and the service curve starting at (x, y). + */ +static void +rtsc_min(struct runtime_sc *rtsc, struct internal_sc *isc, u64 x, u64 y) +{ + u64 y1, y2, dx, dy; + u32 dsm; + + if (isc->sm1 <= isc->sm2) { + /* service curve is convex */ + y1 = rtsc_x2y(rtsc, x); + if (y1 < y) + /* the current rtsc is smaller */ + return; + rtsc->x = x; + rtsc->y = y; + return; + } + + /* + * service curve is concave + * compute the two y values of the current rtsc + * y1: at x + * y2: at (x + dx) + */ + y1 = rtsc_x2y(rtsc, x); + if (y1 <= y) { + /* rtsc is below isc, no change to rtsc */ + return; + } + + y2 = rtsc_x2y(rtsc, x + isc->dx); + if (y2 >= y + isc->dy) { + /* rtsc is above isc, replace rtsc by isc */ + rtsc->x = x; + rtsc->y = y; + rtsc->dx = isc->dx; + rtsc->dy = isc->dy; + return; + } + + /* + * the two curves intersect + * compute the offsets (dx, dy) using the reverse + * function of seg_x2y() + * seg_x2y(dx, sm1) == seg_x2y(dx, sm2) + (y1 - y) + */ + dx = (y1 - y) << SM_SHIFT; + dsm = isc->sm1 - isc->sm2; + do_div(dx, dsm); + /* + * check if (x, y1) belongs to the 1st segment of rtsc. + * if so, add the offset. + */ + if (rtsc->x + rtsc->dx > x) + dx += rtsc->x + rtsc->dx - x; + dy = seg_x2y(dx, isc->sm1); + + rtsc->x = x; + rtsc->y = y; + rtsc->dx = dx; + rtsc->dy = dy; +} + +static void +init_ed(struct hfsc_class *cl, unsigned int next_len) +{ + u64 cur_time = psched_get_time(); + + /* update the deadline curve */ + rtsc_min(&cl->cl_deadline, &cl->cl_rsc, cur_time, cl->cl_cumul); + + /* + * update the eligible curve. + * for concave, it is equal to the deadline curve. + * for convex, it is a linear curve with slope m2. + */ + cl->cl_eligible = cl->cl_deadline; + if (cl->cl_rsc.sm1 <= cl->cl_rsc.sm2) { + cl->cl_eligible.dx = 0; + cl->cl_eligible.dy = 0; + } + + /* compute e and d */ + cl->cl_e = rtsc_y2x(&cl->cl_eligible, cl->cl_cumul); + cl->cl_d = rtsc_y2x(&cl->cl_deadline, cl->cl_cumul + next_len); + + eltree_insert(cl); +} + +static void +update_ed(struct hfsc_class *cl, unsigned int next_len) +{ + cl->cl_e = rtsc_y2x(&cl->cl_eligible, cl->cl_cumul); + cl->cl_d = rtsc_y2x(&cl->cl_deadline, cl->cl_cumul + next_len); + + eltree_update(cl); +} + +static inline void +update_d(struct hfsc_class *cl, unsigned int next_len) +{ + cl->cl_d = rtsc_y2x(&cl->cl_deadline, cl->cl_cumul + next_len); +} + +static inline void +update_cfmin(struct hfsc_class *cl) +{ + struct rb_node *n = rb_first(&cl->cf_tree); + struct hfsc_class *p; + + if (n == NULL) { + cl->cl_cfmin = 0; + return; + } + p = rb_entry(n, struct hfsc_class, cf_node); + cl->cl_cfmin = p->cl_f; +} + +static void +init_vf(struct hfsc_class *cl, unsigned int len) +{ + struct hfsc_class *max_cl; + struct rb_node *n; + u64 vt, f, cur_time; + int go_active; + + cur_time = 0; + go_active = 1; + for (; cl->cl_parent != NULL; cl = cl->cl_parent) { + if (go_active && cl->cl_nactive++ == 0) + go_active = 1; + else + go_active = 0; + + if (go_active) { + n = rb_last(&cl->cl_parent->vt_tree); + if (n != NULL) { + max_cl = rb_entry(n, struct hfsc_class, vt_node); + /* + * set vt to the average of the min and max + * classes. if the parent's period didn't + * change, don't decrease vt of the class. + */ + vt = max_cl->cl_vt; + if (cl->cl_parent->cl_cvtmin != 0) + vt = (cl->cl_parent->cl_cvtmin + vt)/2; + + if (cl->cl_parent->cl_vtperiod != + cl->cl_parentperiod || vt > cl->cl_vt) + cl->cl_vt = vt; + } else { + /* + * first child for a new parent backlog period. + * initialize cl_vt to the highest value seen + * among the siblings. this is analogous to + * what cur_time would provide in realtime case. + */ + cl->cl_vt = cl->cl_parent->cl_cvtoff; + cl->cl_parent->cl_cvtmin = 0; + } + + /* update the virtual curve */ + rtsc_min(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vt, cl->cl_total); + cl->cl_vtadj = 0; + + cl->cl_vtperiod++; /* increment vt period */ + cl->cl_parentperiod = cl->cl_parent->cl_vtperiod; + if (cl->cl_parent->cl_nactive == 0) + cl->cl_parentperiod++; + cl->cl_f = 0; + + vttree_insert(cl); + cftree_insert(cl); + + if (cl->cl_flags & HFSC_USC) { + /* class has upper limit curve */ + if (cur_time == 0) + cur_time = psched_get_time(); + + /* update the ulimit curve */ + rtsc_min(&cl->cl_ulimit, &cl->cl_usc, cur_time, + cl->cl_total); + /* compute myf */ + cl->cl_myf = rtsc_y2x(&cl->cl_ulimit, + cl->cl_total); + } + } + + f = max(cl->cl_myf, cl->cl_cfmin); + if (f != cl->cl_f) { + cl->cl_f = f; + cftree_update(cl); + } + update_cfmin(cl->cl_parent); + } +} + +static void +update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) +{ + u64 f; /* , myf_bound, delta; */ + int go_passive = 0; + + if (cl->qdisc->q.qlen == 0 && cl->cl_flags & HFSC_FSC) + go_passive = 1; + + for (; cl->cl_parent != NULL; cl = cl->cl_parent) { + cl->cl_total += len; + + if (!(cl->cl_flags & HFSC_FSC) || cl->cl_nactive == 0) + continue; + + if (go_passive && --cl->cl_nactive == 0) + go_passive = 1; + else + go_passive = 0; + + /* update vt */ + cl->cl_vt = rtsc_y2x(&cl->cl_virtual, cl->cl_total) + cl->cl_vtadj; + + /* + * if vt of the class is smaller than cvtmin, + * the class was skipped in the past due to non-fit. + * if so, we need to adjust vtadj. + */ + if (cl->cl_vt < cl->cl_parent->cl_cvtmin) { + cl->cl_vtadj += cl->cl_parent->cl_cvtmin - cl->cl_vt; + cl->cl_vt = cl->cl_parent->cl_cvtmin; + } + + if (go_passive) { + /* no more active child, going passive */ + + /* update cvtoff of the parent class */ + if (cl->cl_vt > cl->cl_parent->cl_cvtoff) + cl->cl_parent->cl_cvtoff = cl->cl_vt; + + /* remove this class from the vt tree */ + vttree_remove(cl); + + cftree_remove(cl); + update_cfmin(cl->cl_parent); + + continue; + } + + /* update the vt tree */ + vttree_update(cl); + + /* update f */ + if (cl->cl_flags & HFSC_USC) { + cl->cl_myf = rtsc_y2x(&cl->cl_ulimit, cl->cl_total); +#if 0 + cl->cl_myf = cl->cl_myfadj + rtsc_y2x(&cl->cl_ulimit, + cl->cl_total); + /* + * This code causes classes to stay way under their + * limit when multiple classes are used at gigabit + * speed. needs investigation. -kaber + */ + /* + * if myf lags behind by more than one clock tick + * from the current time, adjust myfadj to prevent + * a rate-limited class from going greedy. + * in a steady state under rate-limiting, myf + * fluctuates within one clock tick. + */ + myf_bound = cur_time - PSCHED_JIFFIE2US(1); + if (cl->cl_myf < myf_bound) { + delta = cur_time - cl->cl_myf; + cl->cl_myfadj += delta; + cl->cl_myf += delta; + } +#endif + } + + f = max(cl->cl_myf, cl->cl_cfmin); + if (f != cl->cl_f) { + cl->cl_f = f; + cftree_update(cl); + update_cfmin(cl->cl_parent); + } + } +} + +static unsigned int +qdisc_peek_len(struct Qdisc *sch) +{ + struct sk_buff *skb; + unsigned int len; + + skb = sch->ops->peek(sch); + if (unlikely(skb == NULL)) { + qdisc_warn_nonwc("qdisc_peek_len", sch); + return 0; + } + len = qdisc_pkt_len(skb); + + return len; +} + +static void +hfsc_purge_queue(struct Qdisc *sch, struct hfsc_class *cl) +{ + unsigned int len = cl->qdisc->q.qlen; + unsigned int backlog = cl->qdisc->qstats.backlog; + + qdisc_reset(cl->qdisc); + qdisc_tree_reduce_backlog(cl->qdisc, len, backlog); +} + +static void +hfsc_adjust_levels(struct hfsc_class *cl) +{ + struct hfsc_class *p; + unsigned int level; + + do { + level = 0; + list_for_each_entry(p, &cl->children, siblings) { + if (p->level >= level) + level = p->level + 1; + } + cl->level = level; + } while ((cl = cl->cl_parent) != NULL); +} + +static inline struct hfsc_class * +hfsc_find_class(u32 classid, struct Qdisc *sch) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct Qdisc_class_common *clc; + + clc = qdisc_class_find(&q->clhash, classid); + if (clc == NULL) + return NULL; + return container_of(clc, struct hfsc_class, cl_common); +} + +static void +hfsc_change_rsc(struct hfsc_class *cl, struct tc_service_curve *rsc, + u64 cur_time) +{ + sc2isc(rsc, &cl->cl_rsc); + rtsc_init(&cl->cl_deadline, &cl->cl_rsc, cur_time, cl->cl_cumul); + cl->cl_eligible = cl->cl_deadline; + if (cl->cl_rsc.sm1 <= cl->cl_rsc.sm2) { + cl->cl_eligible.dx = 0; + cl->cl_eligible.dy = 0; + } + cl->cl_flags |= HFSC_RSC; +} + +static void +hfsc_change_fsc(struct hfsc_class *cl, struct tc_service_curve *fsc) +{ + sc2isc(fsc, &cl->cl_fsc); + rtsc_init(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vt, cl->cl_total); + cl->cl_flags |= HFSC_FSC; +} + +static void +hfsc_change_usc(struct hfsc_class *cl, struct tc_service_curve *usc, + u64 cur_time) +{ + sc2isc(usc, &cl->cl_usc); + rtsc_init(&cl->cl_ulimit, &cl->cl_usc, cur_time, cl->cl_total); + cl->cl_flags |= HFSC_USC; +} + +static const struct nla_policy hfsc_policy[TCA_HFSC_MAX + 1] = { + [TCA_HFSC_RSC] = { .len = sizeof(struct tc_service_curve) }, + [TCA_HFSC_FSC] = { .len = sizeof(struct tc_service_curve) }, + [TCA_HFSC_USC] = { .len = sizeof(struct tc_service_curve) }, +}; + +static int +hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, + struct nlattr **tca, unsigned long *arg, + struct netlink_ext_ack *extack) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl = (struct hfsc_class *)*arg; + struct hfsc_class *parent = NULL; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_HFSC_MAX + 1]; + struct tc_service_curve *rsc = NULL, *fsc = NULL, *usc = NULL; + u64 cur_time; + int err; + + if (opt == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_HFSC_MAX, opt, hfsc_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_HFSC_RSC]) { + rsc = nla_data(tb[TCA_HFSC_RSC]); + if (rsc->m1 == 0 && rsc->m2 == 0) + rsc = NULL; + } + + if (tb[TCA_HFSC_FSC]) { + fsc = nla_data(tb[TCA_HFSC_FSC]); + if (fsc->m1 == 0 && fsc->m2 == 0) + fsc = NULL; + } + + if (tb[TCA_HFSC_USC]) { + usc = nla_data(tb[TCA_HFSC_USC]); + if (usc->m1 == 0 && usc->m2 == 0) + usc = NULL; + } + + if (cl != NULL) { + int old_flags; + + if (parentid) { + if (cl->cl_parent && + cl->cl_parent->cl_common.classid != parentid) + return -EINVAL; + if (cl->cl_parent == NULL && parentid != TC_H_ROOT) + return -EINVAL; + } + cur_time = psched_get_time(); + + if (tca[TCA_RATE]) { + err = gen_replace_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) + return err; + } + + sch_tree_lock(sch); + old_flags = cl->cl_flags; + + if (rsc != NULL) + hfsc_change_rsc(cl, rsc, cur_time); + if (fsc != NULL) + hfsc_change_fsc(cl, fsc); + if (usc != NULL) + hfsc_change_usc(cl, usc, cur_time); + + if (cl->qdisc->q.qlen != 0) { + int len = qdisc_peek_len(cl->qdisc); + + if (cl->cl_flags & HFSC_RSC) { + if (old_flags & HFSC_RSC) + update_ed(cl, len); + else + init_ed(cl, len); + } + + if (cl->cl_flags & HFSC_FSC) { + if (old_flags & HFSC_FSC) + update_vf(cl, 0, cur_time); + else + init_vf(cl, len); + } + } + sch_tree_unlock(sch); + + return 0; + } + + if (parentid == TC_H_ROOT) + return -EEXIST; + + parent = &q->root; + if (parentid) { + parent = hfsc_find_class(parentid, sch); + if (parent == NULL) + return -ENOENT; + } + + if (classid == 0 || TC_H_MAJ(classid ^ sch->handle) != 0) + return -EINVAL; + if (hfsc_find_class(classid, sch)) + return -EEXIST; + + if (rsc == NULL && fsc == NULL) + return -EINVAL; + + cl = kzalloc(sizeof(struct hfsc_class), GFP_KERNEL); + if (cl == NULL) + return -ENOBUFS; + + err = tcf_block_get(&cl->block, &cl->filter_list, sch, extack); + if (err) { + kfree(cl); + return err; + } + + if (tca[TCA_RATE]) { + err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) { + tcf_block_put(cl->block); + kfree(cl); + return err; + } + } + + if (rsc != NULL) + hfsc_change_rsc(cl, rsc, 0); + if (fsc != NULL) + hfsc_change_fsc(cl, fsc); + if (usc != NULL) + hfsc_change_usc(cl, usc, 0); + + cl->cl_common.classid = classid; + cl->sched = q; + cl->cl_parent = parent; + cl->qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + classid, NULL); + if (cl->qdisc == NULL) + cl->qdisc = &noop_qdisc; + else + qdisc_hash_add(cl->qdisc, true); + INIT_LIST_HEAD(&cl->children); + cl->vt_tree = RB_ROOT; + cl->cf_tree = RB_ROOT; + + sch_tree_lock(sch); + qdisc_class_hash_insert(&q->clhash, &cl->cl_common); + list_add_tail(&cl->siblings, &parent->children); + if (parent->level == 0) + hfsc_purge_queue(sch, parent); + hfsc_adjust_levels(parent); + sch_tree_unlock(sch); + + qdisc_class_hash_grow(sch, &q->clhash); + + *arg = (unsigned long)cl; + return 0; +} + +static void +hfsc_destroy_class(struct Qdisc *sch, struct hfsc_class *cl) +{ + struct hfsc_sched *q = qdisc_priv(sch); + + tcf_block_put(cl->block); + qdisc_put(cl->qdisc); + gen_kill_estimator(&cl->rate_est); + if (cl != &q->root) + kfree(cl); +} + +static int +hfsc_delete_class(struct Qdisc *sch, unsigned long arg) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl = (struct hfsc_class *)arg; + + if (cl->level > 0 || cl->filter_cnt > 0 || cl == &q->root) + return -EBUSY; + + sch_tree_lock(sch); + + list_del(&cl->siblings); + hfsc_adjust_levels(cl->cl_parent); + + hfsc_purge_queue(sch, cl); + qdisc_class_hash_remove(&q->clhash, &cl->cl_common); + + sch_tree_unlock(sch); + + hfsc_destroy_class(sch, cl); + return 0; +} + +static struct hfsc_class * +hfsc_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *head, *cl; + struct tcf_result res; + struct tcf_proto *tcf; + int result; + + if (TC_H_MAJ(skb->priority ^ sch->handle) == 0 && + (cl = hfsc_find_class(skb->priority, sch)) != NULL) + if (cl->level == 0) + return cl; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + head = &q->root; + tcf = rcu_dereference_bh(q->root.filter_list); + while (tcf && (result = tcf_classify(skb, tcf, &res, false)) >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + } +#endif + cl = (struct hfsc_class *)res.class; + if (!cl) { + cl = hfsc_find_class(res.classid, sch); + if (!cl) + break; /* filter selected invalid classid */ + if (cl->level >= head->level) + break; /* filter may only point downwards */ + } + + if (cl->level == 0) + return cl; /* hit leaf class */ + + /* apply inner filter chain */ + tcf = rcu_dereference_bh(cl->filter_list); + head = cl; + } + + /* classification failed, try default class */ + cl = hfsc_find_class(TC_H_MAKE(TC_H_MAJ(sch->handle), q->defcls), sch); + if (cl == NULL || cl->level > 0) + return NULL; + + return cl; +} + +static int +hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct hfsc_class *cl = (struct hfsc_class *)arg; + + if (cl->level > 0) + return -EINVAL; + if (new == NULL) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + cl->cl_common.classid, NULL); + if (new == NULL) + new = &noop_qdisc; + } + + *old = qdisc_replace(sch, new, &cl->qdisc); + return 0; +} + +static struct Qdisc * +hfsc_class_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct hfsc_class *cl = (struct hfsc_class *)arg; + + if (cl->level == 0) + return cl->qdisc; + + return NULL; +} + +static void +hfsc_qlen_notify(struct Qdisc *sch, unsigned long arg) +{ + struct hfsc_class *cl = (struct hfsc_class *)arg; + + /* vttree is now handled in update_vf() so that update_vf(cl, 0, 0) + * needs to be called explicitly to remove a class from vttree. + */ + update_vf(cl, 0, 0); + if (cl->cl_flags & HFSC_RSC) + eltree_remove(cl); +} + +static unsigned long +hfsc_search_class(struct Qdisc *sch, u32 classid) +{ + return (unsigned long)hfsc_find_class(classid, sch); +} + +static unsigned long +hfsc_bind_tcf(struct Qdisc *sch, unsigned long parent, u32 classid) +{ + struct hfsc_class *p = (struct hfsc_class *)parent; + struct hfsc_class *cl = hfsc_find_class(classid, sch); + + if (cl != NULL) { + if (p != NULL && p->level <= cl->level) + return 0; + cl->filter_cnt++; + } + + return (unsigned long)cl; +} + +static void +hfsc_unbind_tcf(struct Qdisc *sch, unsigned long arg) +{ + struct hfsc_class *cl = (struct hfsc_class *)arg; + + cl->filter_cnt--; +} + +static struct tcf_block *hfsc_tcf_block(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl = (struct hfsc_class *)arg; + + if (cl == NULL) + cl = &q->root; + + return cl->block; +} + +static int +hfsc_dump_sc(struct sk_buff *skb, int attr, struct internal_sc *sc) +{ + struct tc_service_curve tsc; + + tsc.m1 = sm2m(sc->sm1); + tsc.d = dx2d(sc->dx); + tsc.m2 = sm2m(sc->sm2); + if (nla_put(skb, attr, sizeof(tsc), &tsc)) + goto nla_put_failure; + + return skb->len; + + nla_put_failure: + return -1; +} + +static int +hfsc_dump_curves(struct sk_buff *skb, struct hfsc_class *cl) +{ + if ((cl->cl_flags & HFSC_RSC) && + (hfsc_dump_sc(skb, TCA_HFSC_RSC, &cl->cl_rsc) < 0)) + goto nla_put_failure; + + if ((cl->cl_flags & HFSC_FSC) && + (hfsc_dump_sc(skb, TCA_HFSC_FSC, &cl->cl_fsc) < 0)) + goto nla_put_failure; + + if ((cl->cl_flags & HFSC_USC) && + (hfsc_dump_sc(skb, TCA_HFSC_USC, &cl->cl_usc) < 0)) + goto nla_put_failure; + + return skb->len; + + nla_put_failure: + return -1; +} + +static int +hfsc_dump_class(struct Qdisc *sch, unsigned long arg, struct sk_buff *skb, + struct tcmsg *tcm) +{ + struct hfsc_class *cl = (struct hfsc_class *)arg; + struct nlattr *nest; + + tcm->tcm_parent = cl->cl_parent ? cl->cl_parent->cl_common.classid : + TC_H_ROOT; + tcm->tcm_handle = cl->cl_common.classid; + if (cl->level == 0) + tcm->tcm_info = cl->qdisc->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + if (hfsc_dump_curves(skb, cl) < 0) + goto nla_put_failure; + return nla_nest_end(skb, nest); + + nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static int +hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg, + struct gnet_dump *d) +{ + struct hfsc_class *cl = (struct hfsc_class *)arg; + struct tc_hfsc_stats xstats; + + cl->qstats.backlog = cl->qdisc->qstats.backlog; + xstats.level = cl->level; + xstats.period = cl->cl_vtperiod; + xstats.work = cl->cl_total; + xstats.rtwork = cl->cl_cumul; + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, NULL, &cl->bstats) < 0 || + gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || + gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->qdisc->q.qlen) < 0) + return -1; + + return gnet_stats_copy_app(d, &xstats, sizeof(xstats)); +} + + + +static void +hfsc_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl; + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], + cl_common.hnode) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, (unsigned long)cl, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static void +hfsc_schedule_watchdog(struct Qdisc *sch) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl; + u64 next_time = 0; + + cl = eltree_get_minel(q); + if (cl) + next_time = cl->cl_e; + if (q->root.cl_cfmin != 0) { + if (next_time == 0 || next_time > q->root.cl_cfmin) + next_time = q->root.cl_cfmin; + } + if (next_time) + qdisc_watchdog_schedule(&q->watchdog, next_time); +} + +static int +hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct tc_hfsc_qopt *qopt; + int err; + + qdisc_watchdog_init(&q->watchdog, sch); + + if (!opt || nla_len(opt) < sizeof(*qopt)) + return -EINVAL; + qopt = nla_data(opt); + + q->defcls = qopt->defcls; + err = qdisc_class_hash_init(&q->clhash); + if (err < 0) + return err; + q->eligible = RB_ROOT; + + err = tcf_block_get(&q->root.block, &q->root.filter_list, sch, extack); + if (err) + return err; + + q->root.cl_common.classid = sch->handle; + q->root.sched = q; + q->root.qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + sch->handle, NULL); + if (q->root.qdisc == NULL) + q->root.qdisc = &noop_qdisc; + else + qdisc_hash_add(q->root.qdisc, true); + INIT_LIST_HEAD(&q->root.children); + q->root.vt_tree = RB_ROOT; + q->root.cf_tree = RB_ROOT; + + qdisc_class_hash_insert(&q->clhash, &q->root.cl_common); + qdisc_class_hash_grow(sch, &q->clhash); + + return 0; +} + +static int +hfsc_change_qdisc(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct tc_hfsc_qopt *qopt; + + if (opt == NULL || nla_len(opt) < sizeof(*qopt)) + return -EINVAL; + qopt = nla_data(opt); + + sch_tree_lock(sch); + q->defcls = qopt->defcls; + sch_tree_unlock(sch); + + return 0; +} + +static void +hfsc_reset_class(struct hfsc_class *cl) +{ + cl->cl_total = 0; + cl->cl_cumul = 0; + cl->cl_d = 0; + cl->cl_e = 0; + cl->cl_vt = 0; + cl->cl_vtadj = 0; + cl->cl_cvtmin = 0; + cl->cl_cvtoff = 0; + cl->cl_vtperiod = 0; + cl->cl_parentperiod = 0; + cl->cl_f = 0; + cl->cl_myf = 0; + cl->cl_cfmin = 0; + cl->cl_nactive = 0; + + cl->vt_tree = RB_ROOT; + cl->cf_tree = RB_ROOT; + qdisc_reset(cl->qdisc); + + if (cl->cl_flags & HFSC_RSC) + rtsc_init(&cl->cl_deadline, &cl->cl_rsc, 0, 0); + if (cl->cl_flags & HFSC_FSC) + rtsc_init(&cl->cl_virtual, &cl->cl_fsc, 0, 0); + if (cl->cl_flags & HFSC_USC) + rtsc_init(&cl->cl_ulimit, &cl->cl_usc, 0, 0); +} + +static void +hfsc_reset_qdisc(struct Qdisc *sch) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl; + unsigned int i; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], cl_common.hnode) + hfsc_reset_class(cl); + } + q->eligible = RB_ROOT; + qdisc_watchdog_cancel(&q->watchdog); + sch->qstats.backlog = 0; + sch->q.qlen = 0; +} + +static void +hfsc_destroy_qdisc(struct Qdisc *sch) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hlist_node *next; + struct hfsc_class *cl; + unsigned int i; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], cl_common.hnode) { + tcf_block_put(cl->block); + cl->block = NULL; + } + } + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], + cl_common.hnode) + hfsc_destroy_class(sch, cl); + } + qdisc_class_hash_destroy(&q->clhash); + qdisc_watchdog_cancel(&q->watchdog); +} + +static int +hfsc_dump_qdisc(struct Qdisc *sch, struct sk_buff *skb) +{ + struct hfsc_sched *q = qdisc_priv(sch); + unsigned char *b = skb_tail_pointer(skb); + struct tc_hfsc_qopt qopt; + + qopt.defcls = q->defcls; + if (nla_put(skb, TCA_OPTIONS, sizeof(qopt), &qopt)) + goto nla_put_failure; + return skb->len; + + nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int +hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) +{ + struct hfsc_class *cl; + int uninitialized_var(err); + + cl = hfsc_classify(skb, sch, &err); + if (cl == NULL) { + if (err & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return err; + } + + err = qdisc_enqueue(skb, cl->qdisc, to_free); + if (unlikely(err != NET_XMIT_SUCCESS)) { + if (net_xmit_drop_count(err)) { + cl->qstats.drops++; + qdisc_qstats_drop(sch); + } + return err; + } + + if (cl->qdisc->q.qlen == 1) { + unsigned int len = qdisc_pkt_len(skb); + + if (cl->cl_flags & HFSC_RSC) + init_ed(cl, len); + if (cl->cl_flags & HFSC_FSC) + init_vf(cl, len); + /* + * If this is the first packet, isolate the head so an eventual + * head drop before the first dequeue operation has no chance + * to invalidate the deadline. + */ + if (cl->cl_flags & HFSC_RSC) + cl->qdisc->ops->peek(cl->qdisc); + + } + + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + + return NET_XMIT_SUCCESS; +} + +static struct sk_buff * +hfsc_dequeue(struct Qdisc *sch) +{ + struct hfsc_sched *q = qdisc_priv(sch); + struct hfsc_class *cl; + struct sk_buff *skb; + u64 cur_time; + unsigned int next_len; + int realtime = 0; + + if (sch->q.qlen == 0) + return NULL; + + cur_time = psched_get_time(); + + /* + * if there are eligible classes, use real-time criteria. + * find the class with the minimum deadline among + * the eligible classes. + */ + cl = eltree_get_mindl(q, cur_time); + if (cl) { + realtime = 1; + } else { + /* + * use link-sharing criteria + * get the class with the minimum vt in the hierarchy + */ + cl = vttree_get_minvt(&q->root, cur_time); + if (cl == NULL) { + qdisc_qstats_overlimit(sch); + hfsc_schedule_watchdog(sch); + return NULL; + } + } + + skb = qdisc_dequeue_peeked(cl->qdisc); + if (skb == NULL) { + qdisc_warn_nonwc("HFSC", cl->qdisc); + return NULL; + } + + bstats_update(&cl->bstats, skb); + update_vf(cl, qdisc_pkt_len(skb), cur_time); + if (realtime) + cl->cl_cumul += qdisc_pkt_len(skb); + + if (cl->cl_flags & HFSC_RSC) { + if (cl->qdisc->q.qlen != 0) { + /* update ed */ + next_len = qdisc_peek_len(cl->qdisc); + if (realtime) + update_ed(cl, next_len); + else + update_d(cl, next_len); + } else { + /* the class becomes passive */ + eltree_remove(cl); + } + } + + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + + return skb; +} + +static const struct Qdisc_class_ops hfsc_class_ops = { + .change = hfsc_change_class, + .delete = hfsc_delete_class, + .graft = hfsc_graft_class, + .leaf = hfsc_class_leaf, + .qlen_notify = hfsc_qlen_notify, + .find = hfsc_search_class, + .bind_tcf = hfsc_bind_tcf, + .unbind_tcf = hfsc_unbind_tcf, + .tcf_block = hfsc_tcf_block, + .dump = hfsc_dump_class, + .dump_stats = hfsc_dump_class_stats, + .walk = hfsc_walk +}; + +static struct Qdisc_ops hfsc_qdisc_ops __read_mostly = { + .id = "hfsc", + .init = hfsc_init_qdisc, + .change = hfsc_change_qdisc, + .reset = hfsc_reset_qdisc, + .destroy = hfsc_destroy_qdisc, + .dump = hfsc_dump_qdisc, + .enqueue = hfsc_enqueue, + .dequeue = hfsc_dequeue, + .peek = qdisc_peek_dequeued, + .cl_ops = &hfsc_class_ops, + .priv_size = sizeof(struct hfsc_sched), + .owner = THIS_MODULE +}; + +static int __init +hfsc_init(void) +{ + return register_qdisc(&hfsc_qdisc_ops); +} + +static void __exit +hfsc_cleanup(void) +{ + unregister_qdisc(&hfsc_qdisc_ops); +} + +MODULE_LICENSE("GPL"); +module_init(hfsc_init); +module_exit(hfsc_cleanup); diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c new file mode 100644 index 000000000..c53b72067 --- /dev/null +++ b/net/sched/sch_hhf.c @@ -0,0 +1,721 @@ +/* net/sched/sch_hhf.c Heavy-Hitter Filter (HHF) + * + * Copyright (C) 2013 Terry Lam <vtlam@google.com> + * Copyright (C) 2013 Nandita Dukkipati <nanditad@google.com> + */ + +#include <linux/jiffies.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/vmalloc.h> +#include <linux/siphash.h> +#include <net/pkt_sched.h> +#include <net/sock.h> + +/* Heavy-Hitter Filter (HHF) + * + * Principles : + * Flows are classified into two buckets: non-heavy-hitter and heavy-hitter + * buckets. Initially, a new flow starts as non-heavy-hitter. Once classified + * as heavy-hitter, it is immediately switched to the heavy-hitter bucket. + * The buckets are dequeued by a Weighted Deficit Round Robin (WDRR) scheduler, + * in which the heavy-hitter bucket is served with less weight. + * In other words, non-heavy-hitters (e.g., short bursts of critical traffic) + * are isolated from heavy-hitters (e.g., persistent bulk traffic) and also have + * higher share of bandwidth. + * + * To capture heavy-hitters, we use the "multi-stage filter" algorithm in the + * following paper: + * [EV02] C. Estan and G. Varghese, "New Directions in Traffic Measurement and + * Accounting", in ACM SIGCOMM, 2002. + * + * Conceptually, a multi-stage filter comprises k independent hash functions + * and k counter arrays. Packets are indexed into k counter arrays by k hash + * functions, respectively. The counters are then increased by the packet sizes. + * Therefore, + * - For a heavy-hitter flow: *all* of its k array counters must be large. + * - For a non-heavy-hitter flow: some of its k array counters can be large + * due to hash collision with other small flows; however, with high + * probability, not *all* k counters are large. + * + * By the design of the multi-stage filter algorithm, the false negative rate + * (heavy-hitters getting away uncaptured) is zero. However, the algorithm is + * susceptible to false positives (non-heavy-hitters mistakenly classified as + * heavy-hitters). + * Therefore, we also implement the following optimizations to reduce false + * positives by avoiding unnecessary increment of the counter values: + * - Optimization O1: once a heavy-hitter is identified, its bytes are not + * accounted in the array counters. This technique is called "shielding" + * in Section 3.3.1 of [EV02]. + * - Optimization O2: conservative update of counters + * (Section 3.3.2 of [EV02]), + * New counter value = max {old counter value, + * smallest counter value + packet bytes} + * + * Finally, we refresh the counters periodically since otherwise the counter + * values will keep accumulating. + * + * Once a flow is classified as heavy-hitter, we also save its per-flow state + * in an exact-matching flow table so that its subsequent packets can be + * dispatched to the heavy-hitter bucket accordingly. + * + * + * At a high level, this qdisc works as follows: + * Given a packet p: + * - If the flow-id of p (e.g., TCP 5-tuple) is already in the exact-matching + * heavy-hitter flow table, denoted table T, then send p to the heavy-hitter + * bucket. + * - Otherwise, forward p to the multi-stage filter, denoted filter F + * + If F decides that p belongs to a non-heavy-hitter flow, then send p + * to the non-heavy-hitter bucket. + * + Otherwise, if F decides that p belongs to a new heavy-hitter flow, + * then set up a new flow entry for the flow-id of p in the table T and + * send p to the heavy-hitter bucket. + * + * In this implementation: + * - T is a fixed-size hash-table with 1024 entries. Hash collision is + * resolved by linked-list chaining. + * - F has four counter arrays, each array containing 1024 32-bit counters. + * That means 4 * 1024 * 32 bits = 16KB of memory. + * - Since each array in F contains 1024 counters, 10 bits are sufficient to + * index into each array. + * Hence, instead of having four hash functions, we chop the 32-bit + * skb-hash into three 10-bit chunks, and the remaining 10-bit chunk is + * computed as XOR sum of those three chunks. + * - We need to clear the counter arrays periodically; however, directly + * memsetting 16KB of memory can lead to cache eviction and unwanted delay. + * So by representing each counter by a valid bit, we only need to reset + * 4K of 1 bit (i.e. 512 bytes) instead of 16KB of memory. + * - The Deficit Round Robin engine is taken from fq_codel implementation + * (net/sched/sch_fq_codel.c). Note that wdrr_bucket corresponds to + * fq_codel_flow in fq_codel implementation. + * + */ + +/* Non-configurable parameters */ +#define HH_FLOWS_CNT 1024 /* number of entries in exact-matching table T */ +#define HHF_ARRAYS_CNT 4 /* number of arrays in multi-stage filter F */ +#define HHF_ARRAYS_LEN 1024 /* number of counters in each array of F */ +#define HHF_BIT_MASK_LEN 10 /* masking 10 bits */ +#define HHF_BIT_MASK 0x3FF /* bitmask of 10 bits */ + +#define WDRR_BUCKET_CNT 2 /* two buckets for Weighted DRR */ +enum wdrr_bucket_idx { + WDRR_BUCKET_FOR_HH = 0, /* bucket id for heavy-hitters */ + WDRR_BUCKET_FOR_NON_HH = 1 /* bucket id for non-heavy-hitters */ +}; + +#define hhf_time_before(a, b) \ + (typecheck(u32, a) && typecheck(u32, b) && ((s32)((a) - (b)) < 0)) + +/* Heavy-hitter per-flow state */ +struct hh_flow_state { + u32 hash_id; /* hash of flow-id (e.g. TCP 5-tuple) */ + u32 hit_timestamp; /* last time heavy-hitter was seen */ + struct list_head flowchain; /* chaining under hash collision */ +}; + +/* Weighted Deficit Round Robin (WDRR) scheduler */ +struct wdrr_bucket { + struct sk_buff *head; + struct sk_buff *tail; + struct list_head bucketchain; + int deficit; +}; + +struct hhf_sched_data { + struct wdrr_bucket buckets[WDRR_BUCKET_CNT]; + siphash_key_t perturbation; /* hash perturbation */ + u32 quantum; /* psched_mtu(qdisc_dev(sch)); */ + u32 drop_overlimit; /* number of times max qdisc packet + * limit was hit + */ + struct list_head *hh_flows; /* table T (currently active HHs) */ + u32 hh_flows_limit; /* max active HH allocs */ + u32 hh_flows_overlimit; /* num of disallowed HH allocs */ + u32 hh_flows_total_cnt; /* total admitted HHs */ + u32 hh_flows_current_cnt; /* total current HHs */ + u32 *hhf_arrays[HHF_ARRAYS_CNT]; /* HH filter F */ + u32 hhf_arrays_reset_timestamp; /* last time hhf_arrays + * was reset + */ + unsigned long *hhf_valid_bits[HHF_ARRAYS_CNT]; /* shadow valid bits + * of hhf_arrays + */ + /* Similar to the "new_flows" vs. "old_flows" concept in fq_codel DRR */ + struct list_head new_buckets; /* list of new buckets */ + struct list_head old_buckets; /* list of old buckets */ + + /* Configurable HHF parameters */ + u32 hhf_reset_timeout; /* interval to reset counter + * arrays in filter F + * (default 40ms) + */ + u32 hhf_admit_bytes; /* counter thresh to classify as + * HH (default 128KB). + * With these default values, + * 128KB / 40ms = 25 Mbps + * i.e., we expect to capture HHs + * sending > 25 Mbps. + */ + u32 hhf_evict_timeout; /* aging threshold to evict idle + * HHs out of table T. This should + * be large enough to avoid + * reordering during HH eviction. + * (default 1s) + */ + u32 hhf_non_hh_weight; /* WDRR weight for non-HHs + * (default 2, + * i.e., non-HH : HH = 2 : 1) + */ +}; + +static u32 hhf_time_stamp(void) +{ + return jiffies; +} + +/* Looks up a heavy-hitter flow in a chaining list of table T. */ +static struct hh_flow_state *seek_list(const u32 hash, + struct list_head *head, + struct hhf_sched_data *q) +{ + struct hh_flow_state *flow, *next; + u32 now = hhf_time_stamp(); + + if (list_empty(head)) + return NULL; + + list_for_each_entry_safe(flow, next, head, flowchain) { + u32 prev = flow->hit_timestamp + q->hhf_evict_timeout; + + if (hhf_time_before(prev, now)) { + /* Delete expired heavy-hitters, but preserve one entry + * to avoid kzalloc() when next time this slot is hit. + */ + if (list_is_last(&flow->flowchain, head)) + return NULL; + list_del(&flow->flowchain); + kfree(flow); + q->hh_flows_current_cnt--; + } else if (flow->hash_id == hash) { + return flow; + } + } + return NULL; +} + +/* Returns a flow state entry for a new heavy-hitter. Either reuses an expired + * entry or dynamically alloc a new entry. + */ +static struct hh_flow_state *alloc_new_hh(struct list_head *head, + struct hhf_sched_data *q) +{ + struct hh_flow_state *flow; + u32 now = hhf_time_stamp(); + + if (!list_empty(head)) { + /* Find an expired heavy-hitter flow entry. */ + list_for_each_entry(flow, head, flowchain) { + u32 prev = flow->hit_timestamp + q->hhf_evict_timeout; + + if (hhf_time_before(prev, now)) + return flow; + } + } + + if (q->hh_flows_current_cnt >= q->hh_flows_limit) { + q->hh_flows_overlimit++; + return NULL; + } + /* Create new entry. */ + flow = kzalloc(sizeof(struct hh_flow_state), GFP_ATOMIC); + if (!flow) + return NULL; + + q->hh_flows_current_cnt++; + INIT_LIST_HEAD(&flow->flowchain); + list_add_tail(&flow->flowchain, head); + + return flow; +} + +/* Assigns packets to WDRR buckets. Implements a multi-stage filter to + * classify heavy-hitters. + */ +static enum wdrr_bucket_idx hhf_classify(struct sk_buff *skb, struct Qdisc *sch) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + u32 tmp_hash, hash; + u32 xorsum, filter_pos[HHF_ARRAYS_CNT], flow_pos; + struct hh_flow_state *flow; + u32 pkt_len, min_hhf_val; + int i; + u32 prev; + u32 now = hhf_time_stamp(); + + /* Reset the HHF counter arrays if this is the right time. */ + prev = q->hhf_arrays_reset_timestamp + q->hhf_reset_timeout; + if (hhf_time_before(prev, now)) { + for (i = 0; i < HHF_ARRAYS_CNT; i++) + bitmap_zero(q->hhf_valid_bits[i], HHF_ARRAYS_LEN); + q->hhf_arrays_reset_timestamp = now; + } + + /* Get hashed flow-id of the skb. */ + hash = skb_get_hash_perturb(skb, &q->perturbation); + + /* Check if this packet belongs to an already established HH flow. */ + flow_pos = hash & HHF_BIT_MASK; + flow = seek_list(hash, &q->hh_flows[flow_pos], q); + if (flow) { /* found its HH flow */ + flow->hit_timestamp = now; + return WDRR_BUCKET_FOR_HH; + } + + /* Now pass the packet through the multi-stage filter. */ + tmp_hash = hash; + xorsum = 0; + for (i = 0; i < HHF_ARRAYS_CNT - 1; i++) { + /* Split the skb_hash into three 10-bit chunks. */ + filter_pos[i] = tmp_hash & HHF_BIT_MASK; + xorsum ^= filter_pos[i]; + tmp_hash >>= HHF_BIT_MASK_LEN; + } + /* The last chunk is computed as XOR sum of other chunks. */ + filter_pos[HHF_ARRAYS_CNT - 1] = xorsum ^ tmp_hash; + + pkt_len = qdisc_pkt_len(skb); + min_hhf_val = ~0U; + for (i = 0; i < HHF_ARRAYS_CNT; i++) { + u32 val; + + if (!test_bit(filter_pos[i], q->hhf_valid_bits[i])) { + q->hhf_arrays[i][filter_pos[i]] = 0; + __set_bit(filter_pos[i], q->hhf_valid_bits[i]); + } + + val = q->hhf_arrays[i][filter_pos[i]] + pkt_len; + if (min_hhf_val > val) + min_hhf_val = val; + } + + /* Found a new HH iff all counter values > HH admit threshold. */ + if (min_hhf_val > q->hhf_admit_bytes) { + /* Just captured a new heavy-hitter. */ + flow = alloc_new_hh(&q->hh_flows[flow_pos], q); + if (!flow) /* memory alloc problem */ + return WDRR_BUCKET_FOR_NON_HH; + flow->hash_id = hash; + flow->hit_timestamp = now; + q->hh_flows_total_cnt++; + + /* By returning without updating counters in q->hhf_arrays, + * we implicitly implement "shielding" (see Optimization O1). + */ + return WDRR_BUCKET_FOR_HH; + } + + /* Conservative update of HHF arrays (see Optimization O2). */ + for (i = 0; i < HHF_ARRAYS_CNT; i++) { + if (q->hhf_arrays[i][filter_pos[i]] < min_hhf_val) + q->hhf_arrays[i][filter_pos[i]] = min_hhf_val; + } + return WDRR_BUCKET_FOR_NON_HH; +} + +/* Removes one skb from head of bucket. */ +static struct sk_buff *dequeue_head(struct wdrr_bucket *bucket) +{ + struct sk_buff *skb = bucket->head; + + bucket->head = skb->next; + skb->next = NULL; + return skb; +} + +/* Tail-adds skb to bucket. */ +static void bucket_add(struct wdrr_bucket *bucket, struct sk_buff *skb) +{ + if (bucket->head == NULL) + bucket->head = skb; + else + bucket->tail->next = skb; + bucket->tail = skb; + skb->next = NULL; +} + +static unsigned int hhf_drop(struct Qdisc *sch, struct sk_buff **to_free) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + struct wdrr_bucket *bucket; + + /* Always try to drop from heavy-hitters first. */ + bucket = &q->buckets[WDRR_BUCKET_FOR_HH]; + if (!bucket->head) + bucket = &q->buckets[WDRR_BUCKET_FOR_NON_HH]; + + if (bucket->head) { + struct sk_buff *skb = dequeue_head(bucket); + + sch->q.qlen--; + qdisc_qstats_backlog_dec(sch, skb); + qdisc_drop(skb, sch, to_free); + } + + /* Return id of the bucket from which the packet was dropped. */ + return bucket - q->buckets; +} + +static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + enum wdrr_bucket_idx idx; + struct wdrr_bucket *bucket; + unsigned int prev_backlog; + + idx = hhf_classify(skb, sch); + + bucket = &q->buckets[idx]; + bucket_add(bucket, skb); + qdisc_qstats_backlog_inc(sch, skb); + + if (list_empty(&bucket->bucketchain)) { + unsigned int weight; + + /* The logic of new_buckets vs. old_buckets is the same as + * new_flows vs. old_flows in the implementation of fq_codel, + * i.e., short bursts of non-HHs should have strict priority. + */ + if (idx == WDRR_BUCKET_FOR_HH) { + /* Always move heavy-hitters to old bucket. */ + weight = 1; + list_add_tail(&bucket->bucketchain, &q->old_buckets); + } else { + weight = q->hhf_non_hh_weight; + list_add_tail(&bucket->bucketchain, &q->new_buckets); + } + bucket->deficit = weight * q->quantum; + } + if (++sch->q.qlen <= sch->limit) + return NET_XMIT_SUCCESS; + + prev_backlog = sch->qstats.backlog; + q->drop_overlimit++; + /* Return Congestion Notification only if we dropped a packet from this + * bucket. + */ + if (hhf_drop(sch, to_free) == idx) + return NET_XMIT_CN; + + /* As we dropped a packet, better let upper stack know this. */ + qdisc_tree_reduce_backlog(sch, 1, prev_backlog - sch->qstats.backlog); + return NET_XMIT_SUCCESS; +} + +static struct sk_buff *hhf_dequeue(struct Qdisc *sch) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb = NULL; + struct wdrr_bucket *bucket; + struct list_head *head; + +begin: + head = &q->new_buckets; + if (list_empty(head)) { + head = &q->old_buckets; + if (list_empty(head)) + return NULL; + } + bucket = list_first_entry(head, struct wdrr_bucket, bucketchain); + + if (bucket->deficit <= 0) { + int weight = (bucket - q->buckets == WDRR_BUCKET_FOR_HH) ? + 1 : q->hhf_non_hh_weight; + + bucket->deficit += weight * q->quantum; + list_move_tail(&bucket->bucketchain, &q->old_buckets); + goto begin; + } + + if (bucket->head) { + skb = dequeue_head(bucket); + sch->q.qlen--; + qdisc_qstats_backlog_dec(sch, skb); + } + + if (!skb) { + /* Force a pass through old_buckets to prevent starvation. */ + if ((head == &q->new_buckets) && !list_empty(&q->old_buckets)) + list_move_tail(&bucket->bucketchain, &q->old_buckets); + else + list_del_init(&bucket->bucketchain); + goto begin; + } + qdisc_bstats_update(sch, skb); + bucket->deficit -= qdisc_pkt_len(skb); + + return skb; +} + +static void hhf_reset(struct Qdisc *sch) +{ + struct sk_buff *skb; + + while ((skb = hhf_dequeue(sch)) != NULL) + rtnl_kfree_skbs(skb, skb); +} + +static void hhf_destroy(struct Qdisc *sch) +{ + int i; + struct hhf_sched_data *q = qdisc_priv(sch); + + for (i = 0; i < HHF_ARRAYS_CNT; i++) { + kvfree(q->hhf_arrays[i]); + kvfree(q->hhf_valid_bits[i]); + } + + if (!q->hh_flows) + return; + + for (i = 0; i < HH_FLOWS_CNT; i++) { + struct hh_flow_state *flow, *next; + struct list_head *head = &q->hh_flows[i]; + + if (list_empty(head)) + continue; + list_for_each_entry_safe(flow, next, head, flowchain) { + list_del(&flow->flowchain); + kfree(flow); + } + } + kvfree(q->hh_flows); +} + +static const struct nla_policy hhf_policy[TCA_HHF_MAX + 1] = { + [TCA_HHF_BACKLOG_LIMIT] = { .type = NLA_U32 }, + [TCA_HHF_QUANTUM] = { .type = NLA_U32 }, + [TCA_HHF_HH_FLOWS_LIMIT] = { .type = NLA_U32 }, + [TCA_HHF_RESET_TIMEOUT] = { .type = NLA_U32 }, + [TCA_HHF_ADMIT_BYTES] = { .type = NLA_U32 }, + [TCA_HHF_EVICT_TIMEOUT] = { .type = NLA_U32 }, + [TCA_HHF_NON_HH_WEIGHT] = { .type = NLA_U32 }, +}; + +static int hhf_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_HHF_MAX + 1]; + unsigned int qlen, prev_backlog; + int err; + u64 non_hh_quantum; + u32 new_quantum = q->quantum; + u32 new_hhf_non_hh_weight = q->hhf_non_hh_weight; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_HHF_MAX, opt, hhf_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_HHF_QUANTUM]) + new_quantum = nla_get_u32(tb[TCA_HHF_QUANTUM]); + + if (tb[TCA_HHF_NON_HH_WEIGHT]) + new_hhf_non_hh_weight = nla_get_u32(tb[TCA_HHF_NON_HH_WEIGHT]); + + non_hh_quantum = (u64)new_quantum * new_hhf_non_hh_weight; + if (non_hh_quantum == 0 || non_hh_quantum > INT_MAX) + return -EINVAL; + + sch_tree_lock(sch); + + if (tb[TCA_HHF_BACKLOG_LIMIT]) + sch->limit = nla_get_u32(tb[TCA_HHF_BACKLOG_LIMIT]); + + q->quantum = new_quantum; + q->hhf_non_hh_weight = new_hhf_non_hh_weight; + + if (tb[TCA_HHF_HH_FLOWS_LIMIT]) + q->hh_flows_limit = nla_get_u32(tb[TCA_HHF_HH_FLOWS_LIMIT]); + + if (tb[TCA_HHF_RESET_TIMEOUT]) { + u32 us = nla_get_u32(tb[TCA_HHF_RESET_TIMEOUT]); + + q->hhf_reset_timeout = usecs_to_jiffies(us); + } + + if (tb[TCA_HHF_ADMIT_BYTES]) + q->hhf_admit_bytes = nla_get_u32(tb[TCA_HHF_ADMIT_BYTES]); + + if (tb[TCA_HHF_EVICT_TIMEOUT]) { + u32 us = nla_get_u32(tb[TCA_HHF_EVICT_TIMEOUT]); + + q->hhf_evict_timeout = usecs_to_jiffies(us); + } + + qlen = sch->q.qlen; + prev_backlog = sch->qstats.backlog; + while (sch->q.qlen > sch->limit) { + struct sk_buff *skb = hhf_dequeue(sch); + + rtnl_kfree_skbs(skb, skb); + } + qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, + prev_backlog - sch->qstats.backlog); + + sch_tree_unlock(sch); + return 0; +} + +static int hhf_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + int i; + + sch->limit = 1000; + q->quantum = psched_mtu(qdisc_dev(sch)); + get_random_bytes(&q->perturbation, sizeof(q->perturbation)); + INIT_LIST_HEAD(&q->new_buckets); + INIT_LIST_HEAD(&q->old_buckets); + + /* Configurable HHF parameters */ + q->hhf_reset_timeout = HZ / 25; /* 40 ms */ + q->hhf_admit_bytes = 131072; /* 128 KB */ + q->hhf_evict_timeout = HZ; /* 1 sec */ + q->hhf_non_hh_weight = 2; + + if (opt) { + int err = hhf_change(sch, opt, extack); + + if (err) + return err; + } + + if (!q->hh_flows) { + /* Initialize heavy-hitter flow table. */ + q->hh_flows = kvcalloc(HH_FLOWS_CNT, sizeof(struct list_head), + GFP_KERNEL); + if (!q->hh_flows) + return -ENOMEM; + for (i = 0; i < HH_FLOWS_CNT; i++) + INIT_LIST_HEAD(&q->hh_flows[i]); + + /* Cap max active HHs at twice len of hh_flows table. */ + q->hh_flows_limit = 2 * HH_FLOWS_CNT; + q->hh_flows_overlimit = 0; + q->hh_flows_total_cnt = 0; + q->hh_flows_current_cnt = 0; + + /* Initialize heavy-hitter filter arrays. */ + for (i = 0; i < HHF_ARRAYS_CNT; i++) { + q->hhf_arrays[i] = kvcalloc(HHF_ARRAYS_LEN, + sizeof(u32), + GFP_KERNEL); + if (!q->hhf_arrays[i]) { + /* Note: hhf_destroy() will be called + * by our caller. + */ + return -ENOMEM; + } + } + q->hhf_arrays_reset_timestamp = hhf_time_stamp(); + + /* Initialize valid bits of heavy-hitter filter arrays. */ + for (i = 0; i < HHF_ARRAYS_CNT; i++) { + q->hhf_valid_bits[i] = kvzalloc(HHF_ARRAYS_LEN / + BITS_PER_BYTE, GFP_KERNEL); + if (!q->hhf_valid_bits[i]) { + /* Note: hhf_destroy() will be called + * by our caller. + */ + return -ENOMEM; + } + } + + /* Initialize Weighted DRR buckets. */ + for (i = 0; i < WDRR_BUCKET_CNT; i++) { + struct wdrr_bucket *bucket = q->buckets + i; + + INIT_LIST_HEAD(&bucket->bucketchain); + } + } + + return 0; +} + +static int hhf_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + + if (nla_put_u32(skb, TCA_HHF_BACKLOG_LIMIT, sch->limit) || + nla_put_u32(skb, TCA_HHF_QUANTUM, q->quantum) || + nla_put_u32(skb, TCA_HHF_HH_FLOWS_LIMIT, q->hh_flows_limit) || + nla_put_u32(skb, TCA_HHF_RESET_TIMEOUT, + jiffies_to_usecs(q->hhf_reset_timeout)) || + nla_put_u32(skb, TCA_HHF_ADMIT_BYTES, q->hhf_admit_bytes) || + nla_put_u32(skb, TCA_HHF_EVICT_TIMEOUT, + jiffies_to_usecs(q->hhf_evict_timeout)) || + nla_put_u32(skb, TCA_HHF_NON_HH_WEIGHT, q->hhf_non_hh_weight)) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + return -1; +} + +static int hhf_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct hhf_sched_data *q = qdisc_priv(sch); + struct tc_hhf_xstats st = { + .drop_overlimit = q->drop_overlimit, + .hh_overlimit = q->hh_flows_overlimit, + .hh_tot_count = q->hh_flows_total_cnt, + .hh_cur_count = q->hh_flows_current_cnt, + }; + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static struct Qdisc_ops hhf_qdisc_ops __read_mostly = { + .id = "hhf", + .priv_size = sizeof(struct hhf_sched_data), + + .enqueue = hhf_enqueue, + .dequeue = hhf_dequeue, + .peek = qdisc_peek_dequeued, + .init = hhf_init, + .reset = hhf_reset, + .destroy = hhf_destroy, + .change = hhf_change, + .dump = hhf_dump, + .dump_stats = hhf_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init hhf_module_init(void) +{ + return register_qdisc(&hhf_qdisc_ops); +} + +static void __exit hhf_module_exit(void) +{ + unregister_qdisc(&hhf_qdisc_ops); +} + +module_init(hhf_module_init) +module_exit(hhf_module_exit) +MODULE_AUTHOR("Terry Lam"); +MODULE_AUTHOR("Nandita Dukkipati"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c new file mode 100644 index 000000000..862a33b9e --- /dev/null +++ b/net/sched/sch_htb.c @@ -0,0 +1,1621 @@ +/* + * net/sched/sch_htb.c Hierarchical token bucket, feed tree version + * + * 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. + * + * Authors: Martin Devera, <devik@cdi.cz> + * + * Credits (in time order) for older HTB versions: + * Stef Coene <stef.coene@docum.org> + * HTB support at LARTC mailing list + * Ondrej Kraus, <krauso@barr.cz> + * found missing INIT_QDISC(htb) + * Vladimir Smelhaus, Aamer Akhter, Bert Hubert + * helped a lot to locate nasty class stall bug + * Andi Kleen, Jamal Hadi, Bert Hubert + * code review and helpful comments on shaping + * Tomasz Wrona, <tw@eter.tym.pl> + * created test case so that I was able to fix nasty bug + * Wilfried Weissmann + * spotted bug in dequeue code and helped with fix + * Jiri Fojtasek + * fixed requeue routine + * and many others. thanks. + */ +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/list.h> +#include <linux/compiler.h> +#include <linux/rbtree.h> +#include <linux/workqueue.h> +#include <linux/slab.h> +#include <net/netlink.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +/* HTB algorithm. + Author: devik@cdi.cz + ======================================================================== + HTB is like TBF with multiple classes. It is also similar to CBQ because + it allows to assign priority to each class in hierarchy. + In fact it is another implementation of Floyd's formal sharing. + + Levels: + Each class is assigned level. Leaf has ALWAYS level 0 and root + classes have level TC_HTB_MAXDEPTH-1. Interior nodes has level + one less than their parent. +*/ + +static int htb_hysteresis __read_mostly = 0; /* whether to use mode hysteresis for speedup */ +#define HTB_VER 0x30011 /* major must be matched with number suplied by TC as version */ + +#if HTB_VER >> 16 != TC_HTB_PROTOVER +#error "Mismatched sch_htb.c and pkt_sch.h" +#endif + +/* Module parameter and sysfs export */ +module_param (htb_hysteresis, int, 0640); +MODULE_PARM_DESC(htb_hysteresis, "Hysteresis mode, less CPU load, less accurate"); + +static int htb_rate_est = 0; /* htb classes have a default rate estimator */ +module_param(htb_rate_est, int, 0640); +MODULE_PARM_DESC(htb_rate_est, "setup a default rate estimator (4sec 16sec) for htb classes"); + +/* used internaly to keep status of single class */ +enum htb_cmode { + HTB_CANT_SEND, /* class can't send and can't borrow */ + HTB_MAY_BORROW, /* class can't send but may borrow */ + HTB_CAN_SEND /* class can send */ +}; + +struct htb_prio { + union { + struct rb_root row; + struct rb_root feed; + }; + struct rb_node *ptr; + /* When class changes from state 1->2 and disconnects from + * parent's feed then we lost ptr value and start from the + * first child again. Here we store classid of the + * last valid ptr (used when ptr is NULL). + */ + u32 last_ptr_id; +}; + +/* interior & leaf nodes; props specific to leaves are marked L: + * To reduce false sharing, place mostly read fields at beginning, + * and mostly written ones at the end. + */ +struct htb_class { + struct Qdisc_class_common common; + struct psched_ratecfg rate; + struct psched_ratecfg ceil; + s64 buffer, cbuffer;/* token bucket depth/rate */ + s64 mbuffer; /* max wait time */ + u32 prio; /* these two are used only by leaves... */ + int quantum; /* but stored for parent-to-leaf return */ + + struct tcf_proto __rcu *filter_list; /* class attached filters */ + struct tcf_block *block; + int filter_cnt; + + int level; /* our level (see above) */ + unsigned int children; + struct htb_class *parent; /* parent class */ + + struct net_rate_estimator __rcu *rate_est; + + /* + * Written often fields + */ + struct gnet_stats_basic_packed bstats; + struct tc_htb_xstats xstats; /* our special stats */ + + /* token bucket parameters */ + s64 tokens, ctokens;/* current number of tokens */ + s64 t_c; /* checkpoint time */ + + union { + struct htb_class_leaf { + int deficit[TC_HTB_MAXDEPTH]; + struct Qdisc *q; + } leaf; + struct htb_class_inner { + struct htb_prio clprio[TC_HTB_NUMPRIO]; + } inner; + } un; + s64 pq_key; + + int prio_activity; /* for which prios are we active */ + enum htb_cmode cmode; /* current mode of the class */ + struct rb_node pq_node; /* node for event queue */ + struct rb_node node[TC_HTB_NUMPRIO]; /* node for self or feed tree */ + + unsigned int drops ____cacheline_aligned_in_smp; + unsigned int overlimits; +}; + +struct htb_level { + struct rb_root wait_pq; + struct htb_prio hprio[TC_HTB_NUMPRIO]; +}; + +struct htb_sched { + struct Qdisc_class_hash clhash; + int defcls; /* class where unclassified flows go to */ + int rate2quantum; /* quant = rate / rate2quantum */ + + /* filters for qdisc itself */ + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + +#define HTB_WARN_TOOMANYEVENTS 0x1 + unsigned int warned; /* only one warning */ + int direct_qlen; + struct work_struct work; + + /* non shaped skbs; let them go directly thru */ + struct qdisc_skb_head direct_queue; + long direct_pkts; + + struct qdisc_watchdog watchdog; + + s64 now; /* cached dequeue time */ + + /* time of nearest event per level (row) */ + s64 near_ev_cache[TC_HTB_MAXDEPTH]; + + int row_mask[TC_HTB_MAXDEPTH]; + + struct htb_level hlevel[TC_HTB_MAXDEPTH]; +}; + +/* find class in global hash table using given handle */ +static inline struct htb_class *htb_find(u32 handle, struct Qdisc *sch) +{ + struct htb_sched *q = qdisc_priv(sch); + struct Qdisc_class_common *clc; + + clc = qdisc_class_find(&q->clhash, handle); + if (clc == NULL) + return NULL; + return container_of(clc, struct htb_class, common); +} + +static unsigned long htb_search(struct Qdisc *sch, u32 handle) +{ + return (unsigned long)htb_find(handle, sch); +} +/** + * htb_classify - classify a packet into class + * + * It returns NULL if the packet should be dropped or -1 if the packet + * should be passed directly thru. In all other cases leaf class is returned. + * We allow direct class selection by classid in priority. The we examine + * filters in qdisc and in inner nodes (if higher filter points to the inner + * node). If we end up with classid MAJOR:0 we enqueue the skb into special + * internal fifo (direct). These packets then go directly thru. If we still + * have no valid leaf we try to use MAJOR:default leaf. It still unsuccessful + * then finish and return direct queue. + */ +#define HTB_DIRECT ((struct htb_class *)-1L) + +static struct htb_class *htb_classify(struct sk_buff *skb, struct Qdisc *sch, + int *qerr) +{ + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl; + struct tcf_result res; + struct tcf_proto *tcf; + int result; + + /* allow to select class by setting skb->priority to valid classid; + * note that nfmark can be used too by attaching filter fw with no + * rules in it + */ + if (skb->priority == sch->handle) + return HTB_DIRECT; /* X:0 (direct flow) selected */ + cl = htb_find(skb->priority, sch); + if (cl) { + if (cl->level == 0) + return cl; + /* Start with inner filter chain if a non-leaf class is selected */ + tcf = rcu_dereference_bh(cl->filter_list); + } else { + tcf = rcu_dereference_bh(q->filter_list); + } + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + while (tcf && (result = tcf_classify(skb, tcf, &res, false)) >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + } +#endif + cl = (void *)res.class; + if (!cl) { + if (res.classid == sch->handle) + return HTB_DIRECT; /* X:0 (direct flow) */ + cl = htb_find(res.classid, sch); + if (!cl) + break; /* filter selected invalid classid */ + } + if (!cl->level) + return cl; /* we hit leaf; return it */ + + /* we have got inner class; apply inner filter chain */ + tcf = rcu_dereference_bh(cl->filter_list); + } + /* classification failed; try to use default class */ + cl = htb_find(TC_H_MAKE(TC_H_MAJ(sch->handle), q->defcls), sch); + if (!cl || cl->level) + return HTB_DIRECT; /* bad default .. this is safe bet */ + return cl; +} + +/** + * htb_add_to_id_tree - adds class to the round robin list + * + * Routine adds class to the list (actually tree) sorted by classid. + * Make sure that class is not already on such list for given prio. + */ +static void htb_add_to_id_tree(struct rb_root *root, + struct htb_class *cl, int prio) +{ + struct rb_node **p = &root->rb_node, *parent = NULL; + + while (*p) { + struct htb_class *c; + parent = *p; + c = rb_entry(parent, struct htb_class, node[prio]); + + if (cl->common.classid > c->common.classid) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&cl->node[prio], parent, p); + rb_insert_color(&cl->node[prio], root); +} + +/** + * htb_add_to_wait_tree - adds class to the event queue with delay + * + * The class is added to priority event queue to indicate that class will + * change its mode in cl->pq_key microseconds. Make sure that class is not + * already in the queue. + */ +static void htb_add_to_wait_tree(struct htb_sched *q, + struct htb_class *cl, s64 delay) +{ + struct rb_node **p = &q->hlevel[cl->level].wait_pq.rb_node, *parent = NULL; + + cl->pq_key = q->now + delay; + if (cl->pq_key == q->now) + cl->pq_key++; + + /* update the nearest event cache */ + if (q->near_ev_cache[cl->level] > cl->pq_key) + q->near_ev_cache[cl->level] = cl->pq_key; + + while (*p) { + struct htb_class *c; + parent = *p; + c = rb_entry(parent, struct htb_class, pq_node); + if (cl->pq_key >= c->pq_key) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&cl->pq_node, parent, p); + rb_insert_color(&cl->pq_node, &q->hlevel[cl->level].wait_pq); +} + +/** + * htb_next_rb_node - finds next node in binary tree + * + * When we are past last key we return NULL. + * Average complexity is 2 steps per call. + */ +static inline void htb_next_rb_node(struct rb_node **n) +{ + *n = rb_next(*n); +} + +/** + * htb_add_class_to_row - add class to its row + * + * The class is added to row at priorities marked in mask. + * It does nothing if mask == 0. + */ +static inline void htb_add_class_to_row(struct htb_sched *q, + struct htb_class *cl, int mask) +{ + q->row_mask[cl->level] |= mask; + while (mask) { + int prio = ffz(~mask); + mask &= ~(1 << prio); + htb_add_to_id_tree(&q->hlevel[cl->level].hprio[prio].row, cl, prio); + } +} + +/* If this triggers, it is a bug in this code, but it need not be fatal */ +static void htb_safe_rb_erase(struct rb_node *rb, struct rb_root *root) +{ + if (RB_EMPTY_NODE(rb)) { + WARN_ON(1); + } else { + rb_erase(rb, root); + RB_CLEAR_NODE(rb); + } +} + + +/** + * htb_remove_class_from_row - removes class from its row + * + * The class is removed from row at priorities marked in mask. + * It does nothing if mask == 0. + */ +static inline void htb_remove_class_from_row(struct htb_sched *q, + struct htb_class *cl, int mask) +{ + int m = 0; + struct htb_level *hlevel = &q->hlevel[cl->level]; + + while (mask) { + int prio = ffz(~mask); + struct htb_prio *hprio = &hlevel->hprio[prio]; + + mask &= ~(1 << prio); + if (hprio->ptr == cl->node + prio) + htb_next_rb_node(&hprio->ptr); + + htb_safe_rb_erase(cl->node + prio, &hprio->row); + if (!hprio->row.rb_node) + m |= 1 << prio; + } + q->row_mask[cl->level] &= ~m; +} + +/** + * htb_activate_prios - creates active classe's feed chain + * + * The class is connected to ancestors and/or appropriate rows + * for priorities it is participating on. cl->cmode must be new + * (activated) mode. It does nothing if cl->prio_activity == 0. + */ +static void htb_activate_prios(struct htb_sched *q, struct htb_class *cl) +{ + struct htb_class *p = cl->parent; + long m, mask = cl->prio_activity; + + while (cl->cmode == HTB_MAY_BORROW && p && mask) { + m = mask; + while (m) { + int prio = ffz(~m); + m &= ~(1 << prio); + + if (p->un.inner.clprio[prio].feed.rb_node) + /* parent already has its feed in use so that + * reset bit in mask as parent is already ok + */ + mask &= ~(1 << prio); + + htb_add_to_id_tree(&p->un.inner.clprio[prio].feed, cl, prio); + } + p->prio_activity |= mask; + cl = p; + p = cl->parent; + + } + if (cl->cmode == HTB_CAN_SEND && mask) + htb_add_class_to_row(q, cl, mask); +} + +/** + * htb_deactivate_prios - remove class from feed chain + * + * cl->cmode must represent old mode (before deactivation). It does + * nothing if cl->prio_activity == 0. Class is removed from all feed + * chains and rows. + */ +static void htb_deactivate_prios(struct htb_sched *q, struct htb_class *cl) +{ + struct htb_class *p = cl->parent; + long m, mask = cl->prio_activity; + + while (cl->cmode == HTB_MAY_BORROW && p && mask) { + m = mask; + mask = 0; + while (m) { + int prio = ffz(~m); + m &= ~(1 << prio); + + if (p->un.inner.clprio[prio].ptr == cl->node + prio) { + /* we are removing child which is pointed to from + * parent feed - forget the pointer but remember + * classid + */ + p->un.inner.clprio[prio].last_ptr_id = cl->common.classid; + p->un.inner.clprio[prio].ptr = NULL; + } + + htb_safe_rb_erase(cl->node + prio, + &p->un.inner.clprio[prio].feed); + + if (!p->un.inner.clprio[prio].feed.rb_node) + mask |= 1 << prio; + } + + p->prio_activity &= ~mask; + cl = p; + p = cl->parent; + + } + if (cl->cmode == HTB_CAN_SEND && mask) + htb_remove_class_from_row(q, cl, mask); +} + +static inline s64 htb_lowater(const struct htb_class *cl) +{ + if (htb_hysteresis) + return cl->cmode != HTB_CANT_SEND ? -cl->cbuffer : 0; + else + return 0; +} +static inline s64 htb_hiwater(const struct htb_class *cl) +{ + if (htb_hysteresis) + return cl->cmode == HTB_CAN_SEND ? -cl->buffer : 0; + else + return 0; +} + + +/** + * htb_class_mode - computes and returns current class mode + * + * It computes cl's mode at time cl->t_c+diff and returns it. If mode + * is not HTB_CAN_SEND then cl->pq_key is updated to time difference + * from now to time when cl will change its state. + * Also it is worth to note that class mode doesn't change simply + * at cl->{c,}tokens == 0 but there can rather be hysteresis of + * 0 .. -cl->{c,}buffer range. It is meant to limit number of + * mode transitions per time unit. The speed gain is about 1/6. + */ +static inline enum htb_cmode +htb_class_mode(struct htb_class *cl, s64 *diff) +{ + s64 toks; + + if ((toks = (cl->ctokens + *diff)) < htb_lowater(cl)) { + *diff = -toks; + return HTB_CANT_SEND; + } + + if ((toks = (cl->tokens + *diff)) >= htb_hiwater(cl)) + return HTB_CAN_SEND; + + *diff = -toks; + return HTB_MAY_BORROW; +} + +/** + * htb_change_class_mode - changes classe's mode + * + * This should be the only way how to change classe's mode under normal + * cirsumstances. Routine will update feed lists linkage, change mode + * and add class to the wait event queue if appropriate. New mode should + * be different from old one and cl->pq_key has to be valid if changing + * to mode other than HTB_CAN_SEND (see htb_add_to_wait_tree). + */ +static void +htb_change_class_mode(struct htb_sched *q, struct htb_class *cl, s64 *diff) +{ + enum htb_cmode new_mode = htb_class_mode(cl, diff); + + if (new_mode == cl->cmode) + return; + + if (new_mode == HTB_CANT_SEND) + cl->overlimits++; + + if (cl->prio_activity) { /* not necessary: speed optimization */ + if (cl->cmode != HTB_CANT_SEND) + htb_deactivate_prios(q, cl); + cl->cmode = new_mode; + if (new_mode != HTB_CANT_SEND) + htb_activate_prios(q, cl); + } else + cl->cmode = new_mode; +} + +/** + * htb_activate - inserts leaf cl into appropriate active feeds + * + * Routine learns (new) priority of leaf and activates feed chain + * for the prio. It can be called on already active leaf safely. + * It also adds leaf into droplist. + */ +static inline void htb_activate(struct htb_sched *q, struct htb_class *cl) +{ + WARN_ON(cl->level || !cl->un.leaf.q || !cl->un.leaf.q->q.qlen); + + if (!cl->prio_activity) { + cl->prio_activity = 1 << cl->prio; + htb_activate_prios(q, cl); + } +} + +/** + * htb_deactivate - remove leaf cl from active feeds + * + * Make sure that leaf is active. In the other words it can't be called + * with non-active leaf. It also removes class from the drop list. + */ +static inline void htb_deactivate(struct htb_sched *q, struct htb_class *cl) +{ + WARN_ON(!cl->prio_activity); + + htb_deactivate_prios(q, cl); + cl->prio_activity = 0; +} + +static void htb_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch, + struct qdisc_skb_head *qh) +{ + struct sk_buff *last = qh->tail; + + if (last) { + skb->next = NULL; + last->next = skb; + qh->tail = skb; + } else { + qh->tail = skb; + qh->head = skb; + } + qh->qlen++; +} + +static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + int uninitialized_var(ret); + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl = htb_classify(skb, sch, &ret); + + if (cl == HTB_DIRECT) { + /* enqueue to helper queue */ + if (q->direct_queue.qlen < q->direct_qlen) { + htb_enqueue_tail(skb, sch, &q->direct_queue); + q->direct_pkts++; + } else { + return qdisc_drop(skb, sch, to_free); + } +#ifdef CONFIG_NET_CLS_ACT + } else if (!cl) { + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; +#endif + } else if ((ret = qdisc_enqueue(skb, cl->un.leaf.q, + to_free)) != NET_XMIT_SUCCESS) { + if (net_xmit_drop_count(ret)) { + qdisc_qstats_drop(sch); + cl->drops++; + } + return ret; + } else { + htb_activate(q, cl); + } + + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + return NET_XMIT_SUCCESS; +} + +static inline void htb_accnt_tokens(struct htb_class *cl, int bytes, s64 diff) +{ + s64 toks = diff + cl->tokens; + + if (toks > cl->buffer) + toks = cl->buffer; + toks -= (s64) psched_l2t_ns(&cl->rate, bytes); + if (toks <= -cl->mbuffer) + toks = 1 - cl->mbuffer; + + cl->tokens = toks; +} + +static inline void htb_accnt_ctokens(struct htb_class *cl, int bytes, s64 diff) +{ + s64 toks = diff + cl->ctokens; + + if (toks > cl->cbuffer) + toks = cl->cbuffer; + toks -= (s64) psched_l2t_ns(&cl->ceil, bytes); + if (toks <= -cl->mbuffer) + toks = 1 - cl->mbuffer; + + cl->ctokens = toks; +} + +/** + * htb_charge_class - charges amount "bytes" to leaf and ancestors + * + * Routine assumes that packet "bytes" long was dequeued from leaf cl + * borrowing from "level". It accounts bytes to ceil leaky bucket for + * leaf and all ancestors and to rate bucket for ancestors at levels + * "level" and higher. It also handles possible change of mode resulting + * from the update. Note that mode can also increase here (MAY_BORROW to + * CAN_SEND) because we can use more precise clock that event queue here. + * In such case we remove class from event queue first. + */ +static void htb_charge_class(struct htb_sched *q, struct htb_class *cl, + int level, struct sk_buff *skb) +{ + int bytes = qdisc_pkt_len(skb); + enum htb_cmode old_mode; + s64 diff; + + while (cl) { + diff = min_t(s64, q->now - cl->t_c, cl->mbuffer); + if (cl->level >= level) { + if (cl->level == level) + cl->xstats.lends++; + htb_accnt_tokens(cl, bytes, diff); + } else { + cl->xstats.borrows++; + cl->tokens += diff; /* we moved t_c; update tokens */ + } + htb_accnt_ctokens(cl, bytes, diff); + cl->t_c = q->now; + + old_mode = cl->cmode; + diff = 0; + htb_change_class_mode(q, cl, &diff); + if (old_mode != cl->cmode) { + if (old_mode != HTB_CAN_SEND) + htb_safe_rb_erase(&cl->pq_node, &q->hlevel[cl->level].wait_pq); + if (cl->cmode != HTB_CAN_SEND) + htb_add_to_wait_tree(q, cl, diff); + } + + /* update basic stats except for leaves which are already updated */ + if (cl->level) + bstats_update(&cl->bstats, skb); + + cl = cl->parent; + } +} + +/** + * htb_do_events - make mode changes to classes at the level + * + * Scans event queue for pending events and applies them. Returns time of + * next pending event (0 for no event in pq, q->now for too many events). + * Note: Applied are events whose have cl->pq_key <= q->now. + */ +static s64 htb_do_events(struct htb_sched *q, const int level, + unsigned long start) +{ + /* don't run for longer than 2 jiffies; 2 is used instead of + * 1 to simplify things when jiffy is going to be incremented + * too soon + */ + unsigned long stop_at = start + 2; + struct rb_root *wait_pq = &q->hlevel[level].wait_pq; + + while (time_before(jiffies, stop_at)) { + struct htb_class *cl; + s64 diff; + struct rb_node *p = rb_first(wait_pq); + + if (!p) + return 0; + + cl = rb_entry(p, struct htb_class, pq_node); + if (cl->pq_key > q->now) + return cl->pq_key; + + htb_safe_rb_erase(p, wait_pq); + diff = min_t(s64, q->now - cl->t_c, cl->mbuffer); + htb_change_class_mode(q, cl, &diff); + if (cl->cmode != HTB_CAN_SEND) + htb_add_to_wait_tree(q, cl, diff); + } + + /* too much load - let's continue after a break for scheduling */ + if (!(q->warned & HTB_WARN_TOOMANYEVENTS)) { + pr_warn("htb: too many events!\n"); + q->warned |= HTB_WARN_TOOMANYEVENTS; + } + + return q->now; +} + +/* Returns class->node+prio from id-tree where classe's id is >= id. NULL + * is no such one exists. + */ +static struct rb_node *htb_id_find_next_upper(int prio, struct rb_node *n, + u32 id) +{ + struct rb_node *r = NULL; + while (n) { + struct htb_class *cl = + rb_entry(n, struct htb_class, node[prio]); + + if (id > cl->common.classid) { + n = n->rb_right; + } else if (id < cl->common.classid) { + r = n; + n = n->rb_left; + } else { + return n; + } + } + return r; +} + +/** + * htb_lookup_leaf - returns next leaf class in DRR order + * + * Find leaf where current feed pointers points to. + */ +static struct htb_class *htb_lookup_leaf(struct htb_prio *hprio, const int prio) +{ + int i; + struct { + struct rb_node *root; + struct rb_node **pptr; + u32 *pid; + } stk[TC_HTB_MAXDEPTH], *sp = stk; + + BUG_ON(!hprio->row.rb_node); + sp->root = hprio->row.rb_node; + sp->pptr = &hprio->ptr; + sp->pid = &hprio->last_ptr_id; + + for (i = 0; i < 65535; i++) { + if (!*sp->pptr && *sp->pid) { + /* ptr was invalidated but id is valid - try to recover + * the original or next ptr + */ + *sp->pptr = + htb_id_find_next_upper(prio, sp->root, *sp->pid); + } + *sp->pid = 0; /* ptr is valid now so that remove this hint as it + * can become out of date quickly + */ + if (!*sp->pptr) { /* we are at right end; rewind & go up */ + *sp->pptr = sp->root; + while ((*sp->pptr)->rb_left) + *sp->pptr = (*sp->pptr)->rb_left; + if (sp > stk) { + sp--; + if (!*sp->pptr) { + WARN_ON(1); + return NULL; + } + htb_next_rb_node(sp->pptr); + } + } else { + struct htb_class *cl; + struct htb_prio *clp; + + cl = rb_entry(*sp->pptr, struct htb_class, node[prio]); + if (!cl->level) + return cl; + clp = &cl->un.inner.clprio[prio]; + (++sp)->root = clp->feed.rb_node; + sp->pptr = &clp->ptr; + sp->pid = &clp->last_ptr_id; + } + } + WARN_ON(1); + return NULL; +} + +/* dequeues packet at given priority and level; call only if + * you are sure that there is active class at prio/level + */ +static struct sk_buff *htb_dequeue_tree(struct htb_sched *q, const int prio, + const int level) +{ + struct sk_buff *skb = NULL; + struct htb_class *cl, *start; + struct htb_level *hlevel = &q->hlevel[level]; + struct htb_prio *hprio = &hlevel->hprio[prio]; + + /* look initial class up in the row */ + start = cl = htb_lookup_leaf(hprio, prio); + + do { +next: + if (unlikely(!cl)) + return NULL; + + /* class can be empty - it is unlikely but can be true if leaf + * qdisc drops packets in enqueue routine or if someone used + * graft operation on the leaf since last dequeue; + * simply deactivate and skip such class + */ + if (unlikely(cl->un.leaf.q->q.qlen == 0)) { + struct htb_class *next; + htb_deactivate(q, cl); + + /* row/level might become empty */ + if ((q->row_mask[level] & (1 << prio)) == 0) + return NULL; + + next = htb_lookup_leaf(hprio, prio); + + if (cl == start) /* fix start if we just deleted it */ + start = next; + cl = next; + goto next; + } + + skb = cl->un.leaf.q->dequeue(cl->un.leaf.q); + if (likely(skb != NULL)) + break; + + qdisc_warn_nonwc("htb", cl->un.leaf.q); + htb_next_rb_node(level ? &cl->parent->un.inner.clprio[prio].ptr: + &q->hlevel[0].hprio[prio].ptr); + cl = htb_lookup_leaf(hprio, prio); + + } while (cl != start); + + if (likely(skb != NULL)) { + bstats_update(&cl->bstats, skb); + cl->un.leaf.deficit[level] -= qdisc_pkt_len(skb); + if (cl->un.leaf.deficit[level] < 0) { + cl->un.leaf.deficit[level] += cl->quantum; + htb_next_rb_node(level ? &cl->parent->un.inner.clprio[prio].ptr : + &q->hlevel[0].hprio[prio].ptr); + } + /* this used to be after charge_class but this constelation + * gives us slightly better performance + */ + if (!cl->un.leaf.q->q.qlen) + htb_deactivate(q, cl); + htb_charge_class(q, cl, level, skb); + } + return skb; +} + +static struct sk_buff *htb_dequeue(struct Qdisc *sch) +{ + struct sk_buff *skb; + struct htb_sched *q = qdisc_priv(sch); + int level; + s64 next_event; + unsigned long start_at; + + /* try to dequeue direct packets as high prio (!) to minimize cpu work */ + skb = __qdisc_dequeue_head(&q->direct_queue); + if (skb != NULL) { +ok: + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + return skb; + } + + if (!sch->q.qlen) + goto fin; + q->now = ktime_get_ns(); + start_at = jiffies; + + next_event = q->now + 5LLU * NSEC_PER_SEC; + + for (level = 0; level < TC_HTB_MAXDEPTH; level++) { + /* common case optimization - skip event handler quickly */ + int m; + s64 event = q->near_ev_cache[level]; + + if (q->now >= event) { + event = htb_do_events(q, level, start_at); + if (!event) + event = q->now + NSEC_PER_SEC; + q->near_ev_cache[level] = event; + } + + if (next_event > event) + next_event = event; + + m = ~q->row_mask[level]; + while (m != (int)(-1)) { + int prio = ffz(m); + + m |= 1 << prio; + skb = htb_dequeue_tree(q, prio, level); + if (likely(skb != NULL)) + goto ok; + } + } + qdisc_qstats_overlimit(sch); + if (likely(next_event > q->now)) + qdisc_watchdog_schedule_ns(&q->watchdog, next_event); + else + schedule_work(&q->work); +fin: + return skb; +} + +/* reset all classes */ +/* always caled under BH & queue lock */ +static void htb_reset(struct Qdisc *sch) +{ + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl; + unsigned int i; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + if (cl->level) + memset(&cl->un.inner, 0, sizeof(cl->un.inner)); + else { + if (cl->un.leaf.q) + qdisc_reset(cl->un.leaf.q); + } + cl->prio_activity = 0; + cl->cmode = HTB_CAN_SEND; + } + } + qdisc_watchdog_cancel(&q->watchdog); + __qdisc_reset_queue(&q->direct_queue); + sch->q.qlen = 0; + sch->qstats.backlog = 0; + memset(q->hlevel, 0, sizeof(q->hlevel)); + memset(q->row_mask, 0, sizeof(q->row_mask)); +} + +static const struct nla_policy htb_policy[TCA_HTB_MAX + 1] = { + [TCA_HTB_PARMS] = { .len = sizeof(struct tc_htb_opt) }, + [TCA_HTB_INIT] = { .len = sizeof(struct tc_htb_glob) }, + [TCA_HTB_CTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, + [TCA_HTB_RTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, + [TCA_HTB_DIRECT_QLEN] = { .type = NLA_U32 }, + [TCA_HTB_RATE64] = { .type = NLA_U64 }, + [TCA_HTB_CEIL64] = { .type = NLA_U64 }, +}; + +static void htb_work_func(struct work_struct *work) +{ + struct htb_sched *q = container_of(work, struct htb_sched, work); + struct Qdisc *sch = q->watchdog.qdisc; + + rcu_read_lock(); + __netif_schedule(qdisc_root(sch)); + rcu_read_unlock(); +} + +static int htb_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct htb_sched *q = qdisc_priv(sch); + struct nlattr *tb[TCA_HTB_MAX + 1]; + struct tc_htb_glob *gopt; + int err; + + qdisc_watchdog_init(&q->watchdog, sch); + INIT_WORK(&q->work, htb_work_func); + + if (!opt) + return -EINVAL; + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + err = nla_parse_nested(tb, TCA_HTB_MAX, opt, htb_policy, NULL); + if (err < 0) + return err; + + if (!tb[TCA_HTB_INIT]) + return -EINVAL; + + gopt = nla_data(tb[TCA_HTB_INIT]); + if (gopt->version != HTB_VER >> 16) + return -EINVAL; + + err = qdisc_class_hash_init(&q->clhash); + if (err < 0) + return err; + + qdisc_skb_head_init(&q->direct_queue); + + if (tb[TCA_HTB_DIRECT_QLEN]) + q->direct_qlen = nla_get_u32(tb[TCA_HTB_DIRECT_QLEN]); + else + q->direct_qlen = qdisc_dev(sch)->tx_queue_len; + + if ((q->rate2quantum = gopt->rate2quantum) < 1) + q->rate2quantum = 1; + q->defcls = gopt->defcls; + + return 0; +} + +static int htb_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct htb_sched *q = qdisc_priv(sch); + struct nlattr *nest; + struct tc_htb_glob gopt; + + /* Its safe to not acquire qdisc lock. As we hold RTNL, + * no change can happen on the qdisc parameters. + */ + + gopt.direct_pkts = q->direct_pkts; + gopt.version = HTB_VER; + gopt.rate2quantum = q->rate2quantum; + gopt.defcls = q->defcls; + gopt.debug = 0; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + if (nla_put(skb, TCA_HTB_INIT, sizeof(gopt), &gopt) || + nla_put_u32(skb, TCA_HTB_DIRECT_QLEN, q->direct_qlen)) + goto nla_put_failure; + + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int htb_dump_class(struct Qdisc *sch, unsigned long arg, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct htb_class *cl = (struct htb_class *)arg; + struct nlattr *nest; + struct tc_htb_opt opt; + + /* Its safe to not acquire qdisc lock. As we hold RTNL, + * no change can happen on the class parameters. + */ + tcm->tcm_parent = cl->parent ? cl->parent->common.classid : TC_H_ROOT; + tcm->tcm_handle = cl->common.classid; + if (!cl->level && cl->un.leaf.q) + tcm->tcm_info = cl->un.leaf.q->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + memset(&opt, 0, sizeof(opt)); + + psched_ratecfg_getrate(&opt.rate, &cl->rate); + opt.buffer = PSCHED_NS2TICKS(cl->buffer); + psched_ratecfg_getrate(&opt.ceil, &cl->ceil); + opt.cbuffer = PSCHED_NS2TICKS(cl->cbuffer); + opt.quantum = cl->quantum; + opt.prio = cl->prio; + opt.level = cl->level; + if (nla_put(skb, TCA_HTB_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + if ((cl->rate.rate_bytes_ps >= (1ULL << 32)) && + nla_put_u64_64bit(skb, TCA_HTB_RATE64, cl->rate.rate_bytes_ps, + TCA_HTB_PAD)) + goto nla_put_failure; + if ((cl->ceil.rate_bytes_ps >= (1ULL << 32)) && + nla_put_u64_64bit(skb, TCA_HTB_CEIL64, cl->ceil.rate_bytes_ps, + TCA_HTB_PAD)) + goto nla_put_failure; + + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int +htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) +{ + struct htb_class *cl = (struct htb_class *)arg; + struct gnet_stats_queue qs = { + .drops = cl->drops, + .overlimits = cl->overlimits, + }; + __u32 qlen = 0; + + if (!cl->level && cl->un.leaf.q) { + qlen = cl->un.leaf.q->q.qlen; + qs.backlog = cl->un.leaf.q->qstats.backlog; + } + cl->xstats.tokens = clamp_t(s64, PSCHED_NS2TICKS(cl->tokens), + INT_MIN, INT_MAX); + cl->xstats.ctokens = clamp_t(s64, PSCHED_NS2TICKS(cl->ctokens), + INT_MIN, INT_MAX); + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || + gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || + gnet_stats_copy_queue(d, NULL, &qs, qlen) < 0) + return -1; + + return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats)); +} + +static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct htb_class *cl = (struct htb_class *)arg; + + if (cl->level) + return -EINVAL; + if (new == NULL && + (new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + cl->common.classid, extack)) == NULL) + return -ENOBUFS; + + *old = qdisc_replace(sch, new, &cl->un.leaf.q); + return 0; +} + +static struct Qdisc *htb_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct htb_class *cl = (struct htb_class *)arg; + return !cl->level ? cl->un.leaf.q : NULL; +} + +static void htb_qlen_notify(struct Qdisc *sch, unsigned long arg) +{ + struct htb_class *cl = (struct htb_class *)arg; + + htb_deactivate(qdisc_priv(sch), cl); +} + +static inline int htb_parent_last_child(struct htb_class *cl) +{ + if (!cl->parent) + /* the root class */ + return 0; + if (cl->parent->children > 1) + /* not the last child */ + return 0; + return 1; +} + +static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl, + struct Qdisc *new_q) +{ + struct htb_class *parent = cl->parent; + + WARN_ON(cl->level || !cl->un.leaf.q || cl->prio_activity); + + if (parent->cmode != HTB_CAN_SEND) + htb_safe_rb_erase(&parent->pq_node, + &q->hlevel[parent->level].wait_pq); + + parent->level = 0; + memset(&parent->un.inner, 0, sizeof(parent->un.inner)); + parent->un.leaf.q = new_q ? new_q : &noop_qdisc; + parent->tokens = parent->buffer; + parent->ctokens = parent->cbuffer; + parent->t_c = ktime_get_ns(); + parent->cmode = HTB_CAN_SEND; +} + +static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl) +{ + if (!cl->level) { + WARN_ON(!cl->un.leaf.q); + qdisc_put(cl->un.leaf.q); + } + gen_kill_estimator(&cl->rate_est); + tcf_block_put(cl->block); + kfree(cl); +} + +static void htb_destroy(struct Qdisc *sch) +{ + struct htb_sched *q = qdisc_priv(sch); + struct hlist_node *next; + struct htb_class *cl; + unsigned int i; + + cancel_work_sync(&q->work); + qdisc_watchdog_cancel(&q->watchdog); + /* This line used to be after htb_destroy_class call below + * and surprisingly it worked in 2.4. But it must precede it + * because filter need its target class alive to be able to call + * unbind_filter on it (without Oops). + */ + tcf_block_put(q->block); + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + tcf_block_put(cl->block); + cl->block = NULL; + } + } + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], + common.hnode) + htb_destroy_class(sch, cl); + } + qdisc_class_hash_destroy(&q->clhash); + __qdisc_reset_queue(&q->direct_queue); +} + +static int htb_delete(struct Qdisc *sch, unsigned long arg) +{ + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl = (struct htb_class *)arg; + struct Qdisc *new_q = NULL; + int last_child = 0; + + /* TODO: why don't allow to delete subtree ? references ? does + * tc subsys guarantee us that in htb_destroy it holds no class + * refs so that we can remove children safely there ? + */ + if (cl->children || cl->filter_cnt) + return -EBUSY; + + if (!cl->level && htb_parent_last_child(cl)) { + new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + cl->parent->common.classid, + NULL); + last_child = 1; + } + + sch_tree_lock(sch); + + if (!cl->level) { + unsigned int qlen = cl->un.leaf.q->q.qlen; + unsigned int backlog = cl->un.leaf.q->qstats.backlog; + + qdisc_reset(cl->un.leaf.q); + qdisc_tree_reduce_backlog(cl->un.leaf.q, qlen, backlog); + } + + /* delete from hash and active; remainder in destroy_class */ + qdisc_class_hash_remove(&q->clhash, &cl->common); + if (cl->parent) + cl->parent->children--; + + if (cl->prio_activity) + htb_deactivate(q, cl); + + if (cl->cmode != HTB_CAN_SEND) + htb_safe_rb_erase(&cl->pq_node, + &q->hlevel[cl->level].wait_pq); + + if (last_child) + htb_parent_to_leaf(q, cl, new_q); + + sch_tree_unlock(sch); + + htb_destroy_class(sch, cl); + return 0; +} + +static int htb_change_class(struct Qdisc *sch, u32 classid, + u32 parentid, struct nlattr **tca, + unsigned long *arg, struct netlink_ext_ack *extack) +{ + int err = -EINVAL; + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl = (struct htb_class *)*arg, *parent; + struct nlattr *opt = tca[TCA_OPTIONS]; + struct nlattr *tb[TCA_HTB_MAX + 1]; + struct tc_htb_opt *hopt; + u64 rate64, ceil64; + int warn = 0; + + /* extract all subattrs from opt attr */ + if (!opt) + goto failure; + + err = nla_parse_nested(tb, TCA_HTB_MAX, opt, htb_policy, NULL); + if (err < 0) + goto failure; + + err = -EINVAL; + if (tb[TCA_HTB_PARMS] == NULL) + goto failure; + + parent = parentid == TC_H_ROOT ? NULL : htb_find(parentid, sch); + + hopt = nla_data(tb[TCA_HTB_PARMS]); + if (!hopt->rate.rate || !hopt->ceil.rate) + goto failure; + + /* Keeping backward compatible with rate_table based iproute2 tc */ + if (hopt->rate.linklayer == TC_LINKLAYER_UNAWARE) + qdisc_put_rtab(qdisc_get_rtab(&hopt->rate, tb[TCA_HTB_RTAB], + NULL)); + + if (hopt->ceil.linklayer == TC_LINKLAYER_UNAWARE) + qdisc_put_rtab(qdisc_get_rtab(&hopt->ceil, tb[TCA_HTB_CTAB], + NULL)); + + if (!cl) { /* new class */ + struct Qdisc *new_q; + int prio; + struct { + struct nlattr nla; + struct gnet_estimator opt; + } est = { + .nla = { + .nla_len = nla_attr_size(sizeof(est.opt)), + .nla_type = TCA_RATE, + }, + .opt = { + /* 4s interval, 16s averaging constant */ + .interval = 2, + .ewma_log = 2, + }, + }; + + /* check for valid classid */ + if (!classid || TC_H_MAJ(classid ^ sch->handle) || + htb_find(classid, sch)) + goto failure; + + /* check maximal depth */ + if (parent && parent->parent && parent->parent->level < 2) { + pr_err("htb: tree is too deep\n"); + goto failure; + } + err = -ENOBUFS; + cl = kzalloc(sizeof(*cl), GFP_KERNEL); + if (!cl) + goto failure; + + err = tcf_block_get(&cl->block, &cl->filter_list, sch, extack); + if (err) { + kfree(cl); + goto failure; + } + if (htb_rate_est || tca[TCA_RATE]) { + err = gen_new_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE] ? : &est.nla); + if (err) { + tcf_block_put(cl->block); + kfree(cl); + goto failure; + } + } + + cl->children = 0; + RB_CLEAR_NODE(&cl->pq_node); + + for (prio = 0; prio < TC_HTB_NUMPRIO; prio++) + RB_CLEAR_NODE(&cl->node[prio]); + + /* create leaf qdisc early because it uses kmalloc(GFP_KERNEL) + * so that can't be used inside of sch_tree_lock + * -- thanks to Karlis Peisenieks + */ + new_q = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + classid, NULL); + sch_tree_lock(sch); + if (parent && !parent->level) { + unsigned int qlen = parent->un.leaf.q->q.qlen; + unsigned int backlog = parent->un.leaf.q->qstats.backlog; + + /* turn parent into inner node */ + qdisc_reset(parent->un.leaf.q); + qdisc_tree_reduce_backlog(parent->un.leaf.q, qlen, backlog); + qdisc_put(parent->un.leaf.q); + if (parent->prio_activity) + htb_deactivate(q, parent); + + /* remove from evt list because of level change */ + if (parent->cmode != HTB_CAN_SEND) { + htb_safe_rb_erase(&parent->pq_node, &q->hlevel[0].wait_pq); + parent->cmode = HTB_CAN_SEND; + } + parent->level = (parent->parent ? parent->parent->level + : TC_HTB_MAXDEPTH) - 1; + memset(&parent->un.inner, 0, sizeof(parent->un.inner)); + } + /* leaf (we) needs elementary qdisc */ + cl->un.leaf.q = new_q ? new_q : &noop_qdisc; + + cl->common.classid = classid; + cl->parent = parent; + + /* set class to be in HTB_CAN_SEND state */ + cl->tokens = PSCHED_TICKS2NS(hopt->buffer); + cl->ctokens = PSCHED_TICKS2NS(hopt->cbuffer); + cl->mbuffer = 60ULL * NSEC_PER_SEC; /* 1min */ + cl->t_c = ktime_get_ns(); + cl->cmode = HTB_CAN_SEND; + + /* attach to the hash list and parent's family */ + qdisc_class_hash_insert(&q->clhash, &cl->common); + if (parent) + parent->children++; + if (cl->un.leaf.q != &noop_qdisc) + qdisc_hash_add(cl->un.leaf.q, true); + } else { + if (tca[TCA_RATE]) { + err = gen_replace_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) + return err; + } + sch_tree_lock(sch); + } + + rate64 = tb[TCA_HTB_RATE64] ? nla_get_u64(tb[TCA_HTB_RATE64]) : 0; + + ceil64 = tb[TCA_HTB_CEIL64] ? nla_get_u64(tb[TCA_HTB_CEIL64]) : 0; + + psched_ratecfg_precompute(&cl->rate, &hopt->rate, rate64); + psched_ratecfg_precompute(&cl->ceil, &hopt->ceil, ceil64); + + /* it used to be a nasty bug here, we have to check that node + * is really leaf before changing cl->un.leaf ! + */ + if (!cl->level) { + u64 quantum = cl->rate.rate_bytes_ps; + + do_div(quantum, q->rate2quantum); + cl->quantum = min_t(u64, quantum, INT_MAX); + + if (!hopt->quantum && cl->quantum < 1000) { + warn = -1; + cl->quantum = 1000; + } + if (!hopt->quantum && cl->quantum > 200000) { + warn = 1; + cl->quantum = 200000; + } + if (hopt->quantum) + cl->quantum = hopt->quantum; + if ((cl->prio = hopt->prio) >= TC_HTB_NUMPRIO) + cl->prio = TC_HTB_NUMPRIO - 1; + } + + cl->buffer = PSCHED_TICKS2NS(hopt->buffer); + cl->cbuffer = PSCHED_TICKS2NS(hopt->cbuffer); + + sch_tree_unlock(sch); + + if (warn) + pr_warn("HTB: quantum of class %X is %s. Consider r2q change.\n", + cl->common.classid, (warn == -1 ? "small" : "big")); + + qdisc_class_hash_grow(sch, &q->clhash); + + *arg = (unsigned long)cl; + return 0; + +failure: + return err; +} + +static struct tcf_block *htb_tcf_block(struct Qdisc *sch, unsigned long arg, + struct netlink_ext_ack *extack) +{ + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl = (struct htb_class *)arg; + + return cl ? cl->block : q->block; +} + +static unsigned long htb_bind_filter(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + struct htb_class *cl = htb_find(classid, sch); + + /*if (cl && !cl->level) return 0; + * The line above used to be there to prevent attaching filters to + * leaves. But at least tc_index filter uses this just to get class + * for other reasons so that we have to allow for it. + * ---- + * 19.6.2002 As Werner explained it is ok - bind filter is just + * another way to "lock" the class - unlike "get" this lock can + * be broken by class during destroy IIUC. + */ + if (cl) + cl->filter_cnt++; + return (unsigned long)cl; +} + +static void htb_unbind_filter(struct Qdisc *sch, unsigned long arg) +{ + struct htb_class *cl = (struct htb_class *)arg; + + if (cl) + cl->filter_cnt--; +} + +static void htb_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct htb_sched *q = qdisc_priv(sch); + struct htb_class *cl; + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, (unsigned long)cl, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static const struct Qdisc_class_ops htb_class_ops = { + .graft = htb_graft, + .leaf = htb_leaf, + .qlen_notify = htb_qlen_notify, + .find = htb_search, + .change = htb_change_class, + .delete = htb_delete, + .walk = htb_walk, + .tcf_block = htb_tcf_block, + .bind_tcf = htb_bind_filter, + .unbind_tcf = htb_unbind_filter, + .dump = htb_dump_class, + .dump_stats = htb_dump_class_stats, +}; + +static struct Qdisc_ops htb_qdisc_ops __read_mostly = { + .cl_ops = &htb_class_ops, + .id = "htb", + .priv_size = sizeof(struct htb_sched), + .enqueue = htb_enqueue, + .dequeue = htb_dequeue, + .peek = qdisc_peek_dequeued, + .init = htb_init, + .reset = htb_reset, + .destroy = htb_destroy, + .dump = htb_dump, + .owner = THIS_MODULE, +}; + +static int __init htb_module_init(void) +{ + return register_qdisc(&htb_qdisc_ops); +} +static void __exit htb_module_exit(void) +{ + unregister_qdisc(&htb_qdisc_ops); +} + +module_init(htb_module_init) +module_exit(htb_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c new file mode 100644 index 000000000..ce3f55259 --- /dev/null +++ b/net/sched/sch_ingress.c @@ -0,0 +1,300 @@ +/* net/sched/sch_ingress.c - Ingress and clsact qdisc + * + * 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. + * + * Authors: Jamal Hadi Salim 1999 + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/list.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> + +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +struct ingress_sched_data { + struct tcf_block *block; + struct tcf_block_ext_info block_info; + struct mini_Qdisc_pair miniqp; +}; + +static struct Qdisc *ingress_leaf(struct Qdisc *sch, unsigned long arg) +{ + return NULL; +} + +static unsigned long ingress_find(struct Qdisc *sch, u32 classid) +{ + return TC_H_MIN(classid) + 1; +} + +static unsigned long ingress_bind_filter(struct Qdisc *sch, + unsigned long parent, u32 classid) +{ + return ingress_find(sch, classid); +} + +static void ingress_unbind_filter(struct Qdisc *sch, unsigned long cl) +{ +} + +static void ingress_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ +} + +static struct tcf_block *ingress_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct ingress_sched_data *q = qdisc_priv(sch); + + return q->block; +} + +static void clsact_chain_head_change(struct tcf_proto *tp_head, void *priv) +{ + struct mini_Qdisc_pair *miniqp = priv; + + mini_qdisc_pair_swap(miniqp, tp_head); +}; + +static void ingress_ingress_block_set(struct Qdisc *sch, u32 block_index) +{ + struct ingress_sched_data *q = qdisc_priv(sch); + + q->block_info.block_index = block_index; +} + +static u32 ingress_ingress_block_get(struct Qdisc *sch) +{ + struct ingress_sched_data *q = qdisc_priv(sch); + + return q->block_info.block_index; +} + +static int ingress_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct ingress_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + + net_inc_ingress_queue(); + + mini_qdisc_pair_init(&q->miniqp, sch, &dev->miniq_ingress); + + q->block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; + q->block_info.chain_head_change = clsact_chain_head_change; + q->block_info.chain_head_change_priv = &q->miniqp; + + return tcf_block_get_ext(&q->block, sch, &q->block_info, extack); +} + +static void ingress_destroy(struct Qdisc *sch) +{ + struct ingress_sched_data *q = qdisc_priv(sch); + + tcf_block_put_ext(q->block, sch, &q->block_info); + net_dec_ingress_queue(); +} + +static int ingress_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static const struct Qdisc_class_ops ingress_class_ops = { + .leaf = ingress_leaf, + .find = ingress_find, + .walk = ingress_walk, + .tcf_block = ingress_tcf_block, + .bind_tcf = ingress_bind_filter, + .unbind_tcf = ingress_unbind_filter, +}; + +static struct Qdisc_ops ingress_qdisc_ops __read_mostly = { + .cl_ops = &ingress_class_ops, + .id = "ingress", + .priv_size = sizeof(struct ingress_sched_data), + .static_flags = TCQ_F_CPUSTATS, + .init = ingress_init, + .destroy = ingress_destroy, + .dump = ingress_dump, + .ingress_block_set = ingress_ingress_block_set, + .ingress_block_get = ingress_ingress_block_get, + .owner = THIS_MODULE, +}; + +struct clsact_sched_data { + struct tcf_block *ingress_block; + struct tcf_block *egress_block; + struct tcf_block_ext_info ingress_block_info; + struct tcf_block_ext_info egress_block_info; + struct mini_Qdisc_pair miniqp_ingress; + struct mini_Qdisc_pair miniqp_egress; +}; + +static unsigned long clsact_find(struct Qdisc *sch, u32 classid) +{ + switch (TC_H_MIN(classid)) { + case TC_H_MIN(TC_H_MIN_INGRESS): + case TC_H_MIN(TC_H_MIN_EGRESS): + return TC_H_MIN(classid); + default: + return 0; + } +} + +static unsigned long clsact_bind_filter(struct Qdisc *sch, + unsigned long parent, u32 classid) +{ + return clsact_find(sch, classid); +} + +static struct tcf_block *clsact_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + + switch (cl) { + case TC_H_MIN(TC_H_MIN_INGRESS): + return q->ingress_block; + case TC_H_MIN(TC_H_MIN_EGRESS): + return q->egress_block; + default: + return NULL; + } +} + +static void clsact_ingress_block_set(struct Qdisc *sch, u32 block_index) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + + q->ingress_block_info.block_index = block_index; +} + +static void clsact_egress_block_set(struct Qdisc *sch, u32 block_index) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + + q->egress_block_info.block_index = block_index; +} + +static u32 clsact_ingress_block_get(struct Qdisc *sch) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + + return q->ingress_block_info.block_index; +} + +static u32 clsact_egress_block_get(struct Qdisc *sch) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + + return q->egress_block_info.block_index; +} + +static int clsact_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + int err; + + net_inc_ingress_queue(); + net_inc_egress_queue(); + + mini_qdisc_pair_init(&q->miniqp_ingress, sch, &dev->miniq_ingress); + + q->ingress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; + q->ingress_block_info.chain_head_change = clsact_chain_head_change; + q->ingress_block_info.chain_head_change_priv = &q->miniqp_ingress; + + err = tcf_block_get_ext(&q->ingress_block, sch, &q->ingress_block_info, + extack); + if (err) + return err; + + mini_qdisc_pair_init(&q->miniqp_egress, sch, &dev->miniq_egress); + + q->egress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS; + q->egress_block_info.chain_head_change = clsact_chain_head_change; + q->egress_block_info.chain_head_change_priv = &q->miniqp_egress; + + return tcf_block_get_ext(&q->egress_block, sch, &q->egress_block_info, extack); +} + +static void clsact_destroy(struct Qdisc *sch) +{ + struct clsact_sched_data *q = qdisc_priv(sch); + + tcf_block_put_ext(q->egress_block, sch, &q->egress_block_info); + tcf_block_put_ext(q->ingress_block, sch, &q->ingress_block_info); + + net_dec_ingress_queue(); + net_dec_egress_queue(); +} + +static const struct Qdisc_class_ops clsact_class_ops = { + .leaf = ingress_leaf, + .find = clsact_find, + .walk = ingress_walk, + .tcf_block = clsact_tcf_block, + .bind_tcf = clsact_bind_filter, + .unbind_tcf = ingress_unbind_filter, +}; + +static struct Qdisc_ops clsact_qdisc_ops __read_mostly = { + .cl_ops = &clsact_class_ops, + .id = "clsact", + .priv_size = sizeof(struct clsact_sched_data), + .static_flags = TCQ_F_CPUSTATS, + .init = clsact_init, + .destroy = clsact_destroy, + .dump = ingress_dump, + .ingress_block_set = clsact_ingress_block_set, + .egress_block_set = clsact_egress_block_set, + .ingress_block_get = clsact_ingress_block_get, + .egress_block_get = clsact_egress_block_get, + .owner = THIS_MODULE, +}; + +static int __init ingress_module_init(void) +{ + int ret; + + ret = register_qdisc(&ingress_qdisc_ops); + if (!ret) { + ret = register_qdisc(&clsact_qdisc_ops); + if (ret) + unregister_qdisc(&ingress_qdisc_ops); + } + + return ret; +} + +static void __exit ingress_module_exit(void) +{ + unregister_qdisc(&ingress_qdisc_ops); + unregister_qdisc(&clsact_qdisc_ops); +} + +module_init(ingress_module_init); +module_exit(ingress_module_exit); + +MODULE_ALIAS("sch_clsact"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c new file mode 100644 index 000000000..0ab13a495 --- /dev/null +++ b/net/sched/sch_mq.c @@ -0,0 +1,314 @@ +/* + * net/sched/sch_mq.c Classful multiqueue dummy scheduler + * + * Copyright (c) 2009 Patrick McHardy <kaber@trash.net> + * + * 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/types.h> +#include <linux/slab.h> +#include <linux/kernel.h> +#include <linux/export.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netlink.h> +#include <net/pkt_cls.h> +#include <net/pkt_sched.h> +#include <net/sch_generic.h> + +struct mq_sched { + struct Qdisc **qdiscs; +}; + +static int mq_offload(struct Qdisc *sch, enum tc_mq_command cmd) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_mq_qopt_offload opt = { + .command = cmd, + .handle = sch->handle, + }; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return -EOPNOTSUPP; + + return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_MQ, &opt); +} + +static void mq_offload_stats(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_mq_qopt_offload opt = { + .command = TC_MQ_STATS, + .handle = sch->handle, + .stats = { + .bstats = &sch->bstats, + .qstats = &sch->qstats, + }, + }; + + if (tc_can_offload(dev) && dev->netdev_ops->ndo_setup_tc) + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_MQ, &opt); +} + +static void mq_destroy(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct mq_sched *priv = qdisc_priv(sch); + unsigned int ntx; + + mq_offload(sch, TC_MQ_DESTROY); + + if (!priv->qdiscs) + return; + for (ntx = 0; ntx < dev->num_tx_queues && priv->qdiscs[ntx]; ntx++) + qdisc_put(priv->qdiscs[ntx]); + kfree(priv->qdiscs); +} + +static int mq_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct net_device *dev = qdisc_dev(sch); + struct mq_sched *priv = qdisc_priv(sch); + struct netdev_queue *dev_queue; + struct Qdisc *qdisc; + unsigned int ntx; + + if (sch->parent != TC_H_ROOT) + return -EOPNOTSUPP; + + if (!netif_is_multiqueue(dev)) + return -EOPNOTSUPP; + + /* pre-allocate qdiscs, attachment can't fail */ + priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]), + GFP_KERNEL); + if (!priv->qdiscs) + return -ENOMEM; + + for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { + dev_queue = netdev_get_tx_queue(dev, ntx); + qdisc = qdisc_create_dflt(dev_queue, get_default_qdisc_ops(dev, ntx), + TC_H_MAKE(TC_H_MAJ(sch->handle), + TC_H_MIN(ntx + 1)), + extack); + if (!qdisc) + return -ENOMEM; + priv->qdiscs[ntx] = qdisc; + qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + } + + sch->flags |= TCQ_F_MQROOT; + + mq_offload(sch, TC_MQ_CREATE); + return 0; +} + +static void mq_attach(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct mq_sched *priv = qdisc_priv(sch); + struct Qdisc *qdisc, *old; + unsigned int ntx; + + for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { + qdisc = priv->qdiscs[ntx]; + old = dev_graft_qdisc(qdisc->dev_queue, qdisc); + if (old) + qdisc_put(old); +#ifdef CONFIG_NET_SCHED + if (ntx < dev->real_num_tx_queues) + qdisc_hash_add(qdisc, false); +#endif + + } + kfree(priv->qdiscs); + priv->qdiscs = NULL; +} + +static void mq_change_real_num_tx(struct Qdisc *sch, unsigned int new_real_tx) +{ +#ifdef CONFIG_NET_SCHED + struct net_device *dev = qdisc_dev(sch); + struct Qdisc *qdisc; + unsigned int i; + + for (i = new_real_tx; i < dev->real_num_tx_queues; i++) { + qdisc = netdev_get_tx_queue(dev, i)->qdisc_sleeping; + /* Only update the default qdiscs we created, + * qdiscs with handles are always hashed. + */ + if (qdisc != &noop_qdisc && !qdisc->handle) + qdisc_hash_del(qdisc); + } + for (i = dev->real_num_tx_queues; i < new_real_tx; i++) { + qdisc = netdev_get_tx_queue(dev, i)->qdisc_sleeping; + if (qdisc != &noop_qdisc && !qdisc->handle) + qdisc_hash_add(qdisc, false); + } +#endif +} + +static int mq_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct net_device *dev = qdisc_dev(sch); + struct Qdisc *qdisc; + unsigned int ntx; + __u32 qlen = 0; + + sch->q.qlen = 0; + memset(&sch->bstats, 0, sizeof(sch->bstats)); + memset(&sch->qstats, 0, sizeof(sch->qstats)); + + /* MQ supports lockless qdiscs. However, statistics accounting needs + * to account for all, none, or a mix of locked and unlocked child + * qdiscs. Percpu stats are added to counters in-band and locking + * qdisc totals are added at end. + */ + for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { + qdisc = netdev_get_tx_queue(dev, ntx)->qdisc_sleeping; + spin_lock_bh(qdisc_lock(qdisc)); + + if (qdisc_is_percpu_stats(qdisc)) { + qlen = qdisc_qlen_sum(qdisc); + __gnet_stats_copy_basic(NULL, &sch->bstats, + qdisc->cpu_bstats, + &qdisc->bstats); + __gnet_stats_copy_queue(&sch->qstats, + qdisc->cpu_qstats, + &qdisc->qstats, qlen); + sch->q.qlen += qlen; + } else { + sch->q.qlen += qdisc->q.qlen; + sch->bstats.bytes += qdisc->bstats.bytes; + sch->bstats.packets += qdisc->bstats.packets; + sch->qstats.qlen += qdisc->qstats.qlen; + sch->qstats.backlog += qdisc->qstats.backlog; + sch->qstats.drops += qdisc->qstats.drops; + sch->qstats.requeues += qdisc->qstats.requeues; + sch->qstats.overlimits += qdisc->qstats.overlimits; + } + + spin_unlock_bh(qdisc_lock(qdisc)); + } + mq_offload_stats(sch); + + return 0; +} + +static struct netdev_queue *mq_queue_get(struct Qdisc *sch, unsigned long cl) +{ + struct net_device *dev = qdisc_dev(sch); + unsigned long ntx = cl - 1; + + if (ntx >= dev->num_tx_queues) + return NULL; + return netdev_get_tx_queue(dev, ntx); +} + +static struct netdev_queue *mq_select_queue(struct Qdisc *sch, + struct tcmsg *tcm) +{ + return mq_queue_get(sch, TC_H_MIN(tcm->tcm_parent)); +} + +static int mq_graft(struct Qdisc *sch, unsigned long cl, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct netdev_queue *dev_queue = mq_queue_get(sch, cl); + struct net_device *dev = qdisc_dev(sch); + + if (dev->flags & IFF_UP) + dev_deactivate(dev); + + *old = dev_graft_qdisc(dev_queue, new); + if (new) + new->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + if (dev->flags & IFF_UP) + dev_activate(dev); + return 0; +} + +static struct Qdisc *mq_leaf(struct Qdisc *sch, unsigned long cl) +{ + struct netdev_queue *dev_queue = mq_queue_get(sch, cl); + + return dev_queue->qdisc_sleeping; +} + +static unsigned long mq_find(struct Qdisc *sch, u32 classid) +{ + unsigned int ntx = TC_H_MIN(classid); + + if (!mq_queue_get(sch, ntx)) + return 0; + return ntx; +} + +static int mq_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct netdev_queue *dev_queue = mq_queue_get(sch, cl); + + tcm->tcm_parent = TC_H_ROOT; + tcm->tcm_handle |= TC_H_MIN(cl); + tcm->tcm_info = dev_queue->qdisc_sleeping->handle; + return 0; +} + +static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct netdev_queue *dev_queue = mq_queue_get(sch, cl); + + sch = dev_queue->qdisc_sleeping; + if (gnet_stats_copy_basic(&sch->running, d, sch->cpu_bstats, + &sch->bstats) < 0 || + gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0) + return -1; + return 0; +} + +static void mq_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct net_device *dev = qdisc_dev(sch); + unsigned int ntx; + + if (arg->stop) + return; + + arg->count = arg->skip; + for (ntx = arg->skip; ntx < dev->num_tx_queues; ntx++) { + if (arg->fn(sch, ntx + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } +} + +static const struct Qdisc_class_ops mq_class_ops = { + .select_queue = mq_select_queue, + .graft = mq_graft, + .leaf = mq_leaf, + .find = mq_find, + .walk = mq_walk, + .dump = mq_dump_class, + .dump_stats = mq_dump_class_stats, +}; + +struct Qdisc_ops mq_qdisc_ops __read_mostly = { + .cl_ops = &mq_class_ops, + .id = "mq", + .priv_size = sizeof(struct mq_sched), + .init = mq_init, + .destroy = mq_destroy, + .attach = mq_attach, + .change_real_num_tx = mq_change_real_num_tx, + .dump = mq_dump, + .owner = THIS_MODULE, +}; diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c new file mode 100644 index 000000000..64d7f876d --- /dev/null +++ b/net/sched/sch_mqprio.c @@ -0,0 +1,675 @@ +/* + * net/sched/sch_mqprio.c + * + * Copyright (c) 2010 John Fastabend <john.r.fastabend@intel.com> + * + * 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/types.h> +#include <linux/slab.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/module.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/sch_generic.h> +#include <net/pkt_cls.h> + +struct mqprio_sched { + struct Qdisc **qdiscs; + u16 mode; + u16 shaper; + int hw_offload; + u32 flags; + u64 min_rate[TC_QOPT_MAX_QUEUE]; + u64 max_rate[TC_QOPT_MAX_QUEUE]; +}; + +static void mqprio_destroy(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct mqprio_sched *priv = qdisc_priv(sch); + unsigned int ntx; + + if (priv->qdiscs) { + for (ntx = 0; + ntx < dev->num_tx_queues && priv->qdiscs[ntx]; + ntx++) + qdisc_put(priv->qdiscs[ntx]); + kfree(priv->qdiscs); + } + + if (priv->hw_offload && dev->netdev_ops->ndo_setup_tc) { + struct tc_mqprio_qopt_offload mqprio = { { 0 } }; + + switch (priv->mode) { + case TC_MQPRIO_MODE_DCB: + case TC_MQPRIO_MODE_CHANNEL: + dev->netdev_ops->ndo_setup_tc(dev, + TC_SETUP_QDISC_MQPRIO, + &mqprio); + break; + default: + return; + } + } else { + netdev_set_num_tc(dev, 0); + } +} + +static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt) +{ + int i, j; + + /* Verify num_tc is not out of max range */ + if (qopt->num_tc > TC_MAX_QUEUE) + return -EINVAL; + + /* Verify priority mapping uses valid tcs */ + for (i = 0; i < TC_BITMASK + 1; i++) { + if (qopt->prio_tc_map[i] >= qopt->num_tc) + return -EINVAL; + } + + /* Limit qopt->hw to maximum supported offload value. Drivers have + * the option of overriding this later if they don't support the a + * given offload type. + */ + if (qopt->hw > TC_MQPRIO_HW_OFFLOAD_MAX) + qopt->hw = TC_MQPRIO_HW_OFFLOAD_MAX; + + /* If hardware offload is requested we will leave it to the device + * to either populate the queue counts itself or to validate the + * provided queue counts. If ndo_setup_tc is not present then + * hardware doesn't support offload and we should return an error. + */ + if (qopt->hw) + return dev->netdev_ops->ndo_setup_tc ? 0 : -EINVAL; + + for (i = 0; i < qopt->num_tc; i++) { + unsigned int last = qopt->offset[i] + qopt->count[i]; + + /* Verify the queue count is in tx range being equal to the + * real_num_tx_queues indicates the last queue is in use. + */ + if (qopt->offset[i] >= dev->real_num_tx_queues || + !qopt->count[i] || + last > dev->real_num_tx_queues) + return -EINVAL; + + /* Verify that the offset and counts do not overlap */ + for (j = i + 1; j < qopt->num_tc; j++) { + if (last > qopt->offset[j]) + return -EINVAL; + } + } + + return 0; +} + +static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = { + [TCA_MQPRIO_MODE] = { .len = sizeof(u16) }, + [TCA_MQPRIO_SHAPER] = { .len = sizeof(u16) }, + [TCA_MQPRIO_MIN_RATE64] = { .type = NLA_NESTED }, + [TCA_MQPRIO_MAX_RATE64] = { .type = NLA_NESTED }, +}; + +static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla, + const struct nla_policy *policy, int len) +{ + int nested_len = nla_len(nla) - NLA_ALIGN(len); + + if (nested_len >= nla_attr_size(0)) + return nla_parse(tb, maxtype, nla_data(nla) + NLA_ALIGN(len), + nested_len, policy, NULL); + + memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1)); + return 0; +} + +static int mqprio_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct net_device *dev = qdisc_dev(sch); + struct mqprio_sched *priv = qdisc_priv(sch); + struct netdev_queue *dev_queue; + struct Qdisc *qdisc; + int i, err = -EOPNOTSUPP; + struct tc_mqprio_qopt *qopt = NULL; + struct nlattr *tb[TCA_MQPRIO_MAX + 1]; + struct nlattr *attr; + int rem; + int len; + + BUILD_BUG_ON(TC_MAX_QUEUE != TC_QOPT_MAX_QUEUE); + BUILD_BUG_ON(TC_BITMASK != TC_QOPT_BITMASK); + + if (sch->parent != TC_H_ROOT) + return -EOPNOTSUPP; + + if (!netif_is_multiqueue(dev)) + return -EOPNOTSUPP; + + /* make certain can allocate enough classids to handle queues */ + if (dev->num_tx_queues >= TC_H_MIN_PRIORITY) + return -ENOMEM; + + if (!opt || nla_len(opt) < sizeof(*qopt)) + return -EINVAL; + + qopt = nla_data(opt); + if (mqprio_parse_opt(dev, qopt)) + return -EINVAL; + + len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt)); + if (len > 0) { + err = parse_attr(tb, TCA_MQPRIO_MAX, opt, mqprio_policy, + sizeof(*qopt)); + if (err < 0) + return err; + + if (!qopt->hw) + return -EINVAL; + + if (tb[TCA_MQPRIO_MODE]) { + priv->flags |= TC_MQPRIO_F_MODE; + priv->mode = *(u16 *)nla_data(tb[TCA_MQPRIO_MODE]); + } + + if (tb[TCA_MQPRIO_SHAPER]) { + priv->flags |= TC_MQPRIO_F_SHAPER; + priv->shaper = *(u16 *)nla_data(tb[TCA_MQPRIO_SHAPER]); + } + + if (tb[TCA_MQPRIO_MIN_RATE64]) { + if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) + return -EINVAL; + i = 0; + nla_for_each_nested(attr, tb[TCA_MQPRIO_MIN_RATE64], + rem) { + if (nla_type(attr) != TCA_MQPRIO_MIN_RATE64) + return -EINVAL; + if (i >= qopt->num_tc) + break; + priv->min_rate[i] = *(u64 *)nla_data(attr); + i++; + } + priv->flags |= TC_MQPRIO_F_MIN_RATE; + } + + if (tb[TCA_MQPRIO_MAX_RATE64]) { + if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) + return -EINVAL; + i = 0; + nla_for_each_nested(attr, tb[TCA_MQPRIO_MAX_RATE64], + rem) { + if (nla_type(attr) != TCA_MQPRIO_MAX_RATE64) + return -EINVAL; + if (i >= qopt->num_tc) + break; + priv->max_rate[i] = *(u64 *)nla_data(attr); + i++; + } + priv->flags |= TC_MQPRIO_F_MAX_RATE; + } + } + + /* pre-allocate qdisc, attachment can't fail */ + priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]), + GFP_KERNEL); + if (!priv->qdiscs) + return -ENOMEM; + + for (i = 0; i < dev->num_tx_queues; i++) { + dev_queue = netdev_get_tx_queue(dev, i); + qdisc = qdisc_create_dflt(dev_queue, + get_default_qdisc_ops(dev, i), + TC_H_MAKE(TC_H_MAJ(sch->handle), + TC_H_MIN(i + 1)), extack); + if (!qdisc) + return -ENOMEM; + + priv->qdiscs[i] = qdisc; + qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + } + + /* If the mqprio options indicate that hardware should own + * the queue mapping then run ndo_setup_tc otherwise use the + * supplied and verified mapping + */ + if (qopt->hw) { + struct tc_mqprio_qopt_offload mqprio = {.qopt = *qopt}; + + switch (priv->mode) { + case TC_MQPRIO_MODE_DCB: + if (priv->shaper != TC_MQPRIO_SHAPER_DCB) + return -EINVAL; + break; + case TC_MQPRIO_MODE_CHANNEL: + mqprio.flags = priv->flags; + if (priv->flags & TC_MQPRIO_F_MODE) + mqprio.mode = priv->mode; + if (priv->flags & TC_MQPRIO_F_SHAPER) + mqprio.shaper = priv->shaper; + if (priv->flags & TC_MQPRIO_F_MIN_RATE) + for (i = 0; i < mqprio.qopt.num_tc; i++) + mqprio.min_rate[i] = priv->min_rate[i]; + if (priv->flags & TC_MQPRIO_F_MAX_RATE) + for (i = 0; i < mqprio.qopt.num_tc; i++) + mqprio.max_rate[i] = priv->max_rate[i]; + break; + default: + return -EINVAL; + } + err = dev->netdev_ops->ndo_setup_tc(dev, + TC_SETUP_QDISC_MQPRIO, + &mqprio); + if (err) + return err; + + priv->hw_offload = mqprio.qopt.hw; + } else { + netdev_set_num_tc(dev, qopt->num_tc); + for (i = 0; i < qopt->num_tc; i++) + netdev_set_tc_queue(dev, i, + qopt->count[i], qopt->offset[i]); + } + + /* Always use supplied priority mappings */ + for (i = 0; i < TC_BITMASK + 1; i++) + netdev_set_prio_tc_map(dev, i, qopt->prio_tc_map[i]); + + sch->flags |= TCQ_F_MQROOT; + return 0; +} + +static void mqprio_attach(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct mqprio_sched *priv = qdisc_priv(sch); + struct Qdisc *qdisc, *old; + unsigned int ntx; + + /* Attach underlying qdisc */ + for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { + qdisc = priv->qdiscs[ntx]; + old = dev_graft_qdisc(qdisc->dev_queue, qdisc); + if (old) + qdisc_put(old); + if (ntx < dev->real_num_tx_queues) + qdisc_hash_add(qdisc, false); + } + kfree(priv->qdiscs); + priv->qdiscs = NULL; +} + +static void mqprio_change_real_num_tx(struct Qdisc *sch, + unsigned int new_real_tx) +{ + struct net_device *dev = qdisc_dev(sch); + struct Qdisc *qdisc; + unsigned int i; + + for (i = new_real_tx; i < dev->real_num_tx_queues; i++) { + qdisc = netdev_get_tx_queue(dev, i)->qdisc_sleeping; + /* Only update the default qdiscs we created, + * qdiscs with handles are always hashed. + */ + if (qdisc != &noop_qdisc && !qdisc->handle) + qdisc_hash_del(qdisc); + } + for (i = dev->real_num_tx_queues; i < new_real_tx; i++) { + qdisc = netdev_get_tx_queue(dev, i)->qdisc_sleeping; + if (qdisc != &noop_qdisc && !qdisc->handle) + qdisc_hash_add(qdisc, false); + } +} + +static struct netdev_queue *mqprio_queue_get(struct Qdisc *sch, + unsigned long cl) +{ + struct net_device *dev = qdisc_dev(sch); + unsigned long ntx = cl - 1; + + if (ntx >= dev->num_tx_queues) + return NULL; + return netdev_get_tx_queue(dev, ntx); +} + +static int mqprio_graft(struct Qdisc *sch, unsigned long cl, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct net_device *dev = qdisc_dev(sch); + struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl); + + if (!dev_queue) + return -EINVAL; + + if (dev->flags & IFF_UP) + dev_deactivate(dev); + + *old = dev_graft_qdisc(dev_queue, new); + + if (new) + new->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; + + if (dev->flags & IFF_UP) + dev_activate(dev); + + return 0; +} + +static int dump_rates(struct mqprio_sched *priv, + struct tc_mqprio_qopt *opt, struct sk_buff *skb) +{ + struct nlattr *nest; + int i; + + if (priv->flags & TC_MQPRIO_F_MIN_RATE) { + nest = nla_nest_start(skb, TCA_MQPRIO_MIN_RATE64); + if (!nest) + goto nla_put_failure; + + for (i = 0; i < opt->num_tc; i++) { + if (nla_put(skb, TCA_MQPRIO_MIN_RATE64, + sizeof(priv->min_rate[i]), + &priv->min_rate[i])) + goto nla_put_failure; + } + nla_nest_end(skb, nest); + } + + if (priv->flags & TC_MQPRIO_F_MAX_RATE) { + nest = nla_nest_start(skb, TCA_MQPRIO_MAX_RATE64); + if (!nest) + goto nla_put_failure; + + for (i = 0; i < opt->num_tc; i++) { + if (nla_put(skb, TCA_MQPRIO_MAX_RATE64, + sizeof(priv->max_rate[i]), + &priv->max_rate[i])) + goto nla_put_failure; + } + nla_nest_end(skb, nest); + } + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct net_device *dev = qdisc_dev(sch); + struct mqprio_sched *priv = qdisc_priv(sch); + struct nlattr *nla = (struct nlattr *)skb_tail_pointer(skb); + struct tc_mqprio_qopt opt = { 0 }; + struct Qdisc *qdisc; + unsigned int ntx, tc; + + sch->q.qlen = 0; + memset(&sch->bstats, 0, sizeof(sch->bstats)); + memset(&sch->qstats, 0, sizeof(sch->qstats)); + + /* MQ supports lockless qdiscs. However, statistics accounting needs + * to account for all, none, or a mix of locked and unlocked child + * qdiscs. Percpu stats are added to counters in-band and locking + * qdisc totals are added at end. + */ + for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { + qdisc = netdev_get_tx_queue(dev, ntx)->qdisc_sleeping; + spin_lock_bh(qdisc_lock(qdisc)); + + if (qdisc_is_percpu_stats(qdisc)) { + __u32 qlen = qdisc_qlen_sum(qdisc); + + __gnet_stats_copy_basic(NULL, &sch->bstats, + qdisc->cpu_bstats, + &qdisc->bstats); + __gnet_stats_copy_queue(&sch->qstats, + qdisc->cpu_qstats, + &qdisc->qstats, qlen); + sch->q.qlen += qlen; + } else { + sch->q.qlen += qdisc->q.qlen; + sch->bstats.bytes += qdisc->bstats.bytes; + sch->bstats.packets += qdisc->bstats.packets; + sch->qstats.backlog += qdisc->qstats.backlog; + sch->qstats.drops += qdisc->qstats.drops; + sch->qstats.requeues += qdisc->qstats.requeues; + sch->qstats.overlimits += qdisc->qstats.overlimits; + } + + spin_unlock_bh(qdisc_lock(qdisc)); + } + + opt.num_tc = netdev_get_num_tc(dev); + memcpy(opt.prio_tc_map, dev->prio_tc_map, sizeof(opt.prio_tc_map)); + opt.hw = priv->hw_offload; + + for (tc = 0; tc < netdev_get_num_tc(dev); tc++) { + opt.count[tc] = dev->tc_to_txq[tc].count; + opt.offset[tc] = dev->tc_to_txq[tc].offset; + } + + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + goto nla_put_failure; + + if ((priv->flags & TC_MQPRIO_F_MODE) && + nla_put_u16(skb, TCA_MQPRIO_MODE, priv->mode)) + goto nla_put_failure; + + if ((priv->flags & TC_MQPRIO_F_SHAPER) && + nla_put_u16(skb, TCA_MQPRIO_SHAPER, priv->shaper)) + goto nla_put_failure; + + if ((priv->flags & TC_MQPRIO_F_MIN_RATE || + priv->flags & TC_MQPRIO_F_MAX_RATE) && + (dump_rates(priv, &opt, skb) != 0)) + goto nla_put_failure; + + return nla_nest_end(skb, nla); +nla_put_failure: + nlmsg_trim(skb, nla); + return -1; +} + +static struct Qdisc *mqprio_leaf(struct Qdisc *sch, unsigned long cl) +{ + struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl); + + if (!dev_queue) + return NULL; + + return dev_queue->qdisc_sleeping; +} + +static unsigned long mqprio_find(struct Qdisc *sch, u32 classid) +{ + struct net_device *dev = qdisc_dev(sch); + unsigned int ntx = TC_H_MIN(classid); + + /* There are essentially two regions here that have valid classid + * values. The first region will have a classid value of 1 through + * num_tx_queues. All of these are backed by actual Qdiscs. + */ + if (ntx < TC_H_MIN_PRIORITY) + return (ntx <= dev->num_tx_queues) ? ntx : 0; + + /* The second region represents the hardware traffic classes. These + * are represented by classid values of TC_H_MIN_PRIORITY through + * TC_H_MIN_PRIORITY + netdev_get_num_tc - 1 + */ + return ((ntx - TC_H_MIN_PRIORITY) < netdev_get_num_tc(dev)) ? ntx : 0; +} + +static int mqprio_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + if (cl < TC_H_MIN_PRIORITY) { + struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl); + struct net_device *dev = qdisc_dev(sch); + int tc = netdev_txq_to_tc(dev, cl - 1); + + tcm->tcm_parent = (tc < 0) ? 0 : + TC_H_MAKE(TC_H_MAJ(sch->handle), + TC_H_MIN(tc + TC_H_MIN_PRIORITY)); + tcm->tcm_info = dev_queue->qdisc_sleeping->handle; + } else { + tcm->tcm_parent = TC_H_ROOT; + tcm->tcm_info = 0; + } + tcm->tcm_handle |= TC_H_MIN(cl); + return 0; +} + +static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) + __releases(d->lock) + __acquires(d->lock) +{ + if (cl >= TC_H_MIN_PRIORITY) { + int i; + __u32 qlen = 0; + struct gnet_stats_queue qstats = {0}; + struct gnet_stats_basic_packed bstats = {0}; + struct net_device *dev = qdisc_dev(sch); + struct netdev_tc_txq tc = dev->tc_to_txq[cl & TC_BITMASK]; + + /* Drop lock here it will be reclaimed before touching + * statistics this is required because the d->lock we + * hold here is the look on dev_queue->qdisc_sleeping + * also acquired below. + */ + if (d->lock) + spin_unlock_bh(d->lock); + + for (i = tc.offset; i < tc.offset + tc.count; i++) { + struct netdev_queue *q = netdev_get_tx_queue(dev, i); + struct Qdisc *qdisc = rtnl_dereference(q->qdisc); + + spin_lock_bh(qdisc_lock(qdisc)); + + if (qdisc_is_percpu_stats(qdisc)) { + qlen = qdisc_qlen_sum(qdisc); + + __gnet_stats_copy_basic(NULL, &bstats, + qdisc->cpu_bstats, + &qdisc->bstats); + __gnet_stats_copy_queue(&qstats, + qdisc->cpu_qstats, + &qdisc->qstats, + qlen); + } else { + qlen += qdisc->q.qlen; + bstats.bytes += qdisc->bstats.bytes; + bstats.packets += qdisc->bstats.packets; + qstats.backlog += qdisc->qstats.backlog; + qstats.drops += qdisc->qstats.drops; + qstats.requeues += qdisc->qstats.requeues; + qstats.overlimits += qdisc->qstats.overlimits; + } + spin_unlock_bh(qdisc_lock(qdisc)); + } + + /* Reclaim root sleeping lock before completing stats */ + if (d->lock) + spin_lock_bh(d->lock); + if (gnet_stats_copy_basic(NULL, d, NULL, &bstats) < 0 || + gnet_stats_copy_queue(d, NULL, &qstats, qlen) < 0) + return -1; + } else { + struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl); + + sch = dev_queue->qdisc_sleeping; + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, + sch->cpu_bstats, &sch->bstats) < 0 || + gnet_stats_copy_queue(d, NULL, + &sch->qstats, sch->q.qlen) < 0) + return -1; + } + return 0; +} + +static void mqprio_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct net_device *dev = qdisc_dev(sch); + unsigned long ntx; + + if (arg->stop) + return; + + /* Walk hierarchy with a virtual class per tc */ + arg->count = arg->skip; + for (ntx = arg->skip; ntx < netdev_get_num_tc(dev); ntx++) { + if (arg->fn(sch, ntx + TC_H_MIN_PRIORITY, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + + /* Pad the values and skip over unused traffic classes */ + if (ntx < TC_MAX_QUEUE) { + arg->count = TC_MAX_QUEUE; + ntx = TC_MAX_QUEUE; + } + + /* Reset offset, sort out remaining per-queue qdiscs */ + for (ntx -= TC_MAX_QUEUE; ntx < dev->num_tx_queues; ntx++) { + if (arg->fn(sch, ntx + 1, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } +} + +static struct netdev_queue *mqprio_select_queue(struct Qdisc *sch, + struct tcmsg *tcm) +{ + return mqprio_queue_get(sch, TC_H_MIN(tcm->tcm_parent)); +} + +static const struct Qdisc_class_ops mqprio_class_ops = { + .graft = mqprio_graft, + .leaf = mqprio_leaf, + .find = mqprio_find, + .walk = mqprio_walk, + .dump = mqprio_dump_class, + .dump_stats = mqprio_dump_class_stats, + .select_queue = mqprio_select_queue, +}; + +static struct Qdisc_ops mqprio_qdisc_ops __read_mostly = { + .cl_ops = &mqprio_class_ops, + .id = "mqprio", + .priv_size = sizeof(struct mqprio_sched), + .init = mqprio_init, + .destroy = mqprio_destroy, + .attach = mqprio_attach, + .change_real_num_tx = mqprio_change_real_num_tx, + .dump = mqprio_dump, + .owner = THIS_MODULE, +}; + +static int __init mqprio_module_init(void) +{ + return register_qdisc(&mqprio_qdisc_ops); +} + +static void __exit mqprio_module_exit(void) +{ + unregister_qdisc(&mqprio_qdisc_ops); +} + +module_init(mqprio_module_init); +module_exit(mqprio_module_exit); + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c new file mode 100644 index 000000000..1c2f9a3ab --- /dev/null +++ b/net/sched/sch_multiq.c @@ -0,0 +1,425 @@ +/* + * Copyright (c) 2008, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, see <http://www.gnu.org/licenses/>. + * + * Author: Alexander Duyck <alexander.h.duyck@intel.com> + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +struct multiq_sched_data { + u16 bands; + u16 max_bands; + u16 curband; + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + struct Qdisc **queues; +}; + + +static struct Qdisc * +multiq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + u32 band; + struct tcf_result res; + struct tcf_proto *fl = rcu_dereference_bh(q->filter_list); + int err; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + err = tcf_classify(skb, fl, &res, false); +#ifdef CONFIG_NET_CLS_ACT + switch (err) { + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + } +#endif + band = skb_get_queue_mapping(skb); + + if (band >= q->bands) + return q->queues[0]; + + return q->queues[band]; +} + +static int +multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct Qdisc *qdisc; + int ret; + + qdisc = multiq_classify(skb, sch, &ret); +#ifdef CONFIG_NET_CLS_ACT + if (qdisc == NULL) { + + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; + } +#endif + + ret = qdisc_enqueue(skb, qdisc, to_free); + if (ret == NET_XMIT_SUCCESS) { + sch->q.qlen++; + return NET_XMIT_SUCCESS; + } + if (net_xmit_drop_count(ret)) + qdisc_qstats_drop(sch); + return ret; +} + +static struct sk_buff *multiq_dequeue(struct Qdisc *sch) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + struct Qdisc *qdisc; + struct sk_buff *skb; + int band; + + for (band = 0; band < q->bands; band++) { + /* cycle through bands to ensure fairness */ + q->curband++; + if (q->curband >= q->bands) + q->curband = 0; + + /* Check that target subqueue is available before + * pulling an skb to avoid head-of-line blocking. + */ + if (!netif_xmit_stopped( + netdev_get_tx_queue(qdisc_dev(sch), q->curband))) { + qdisc = q->queues[q->curband]; + skb = qdisc->dequeue(qdisc); + if (skb) { + qdisc_bstats_update(sch, skb); + sch->q.qlen--; + return skb; + } + } + } + return NULL; + +} + +static struct sk_buff *multiq_peek(struct Qdisc *sch) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + unsigned int curband = q->curband; + struct Qdisc *qdisc; + struct sk_buff *skb; + int band; + + for (band = 0; band < q->bands; band++) { + /* cycle through bands to ensure fairness */ + curband++; + if (curband >= q->bands) + curband = 0; + + /* Check that target subqueue is available before + * pulling an skb to avoid head-of-line blocking. + */ + if (!netif_xmit_stopped( + netdev_get_tx_queue(qdisc_dev(sch), curband))) { + qdisc = q->queues[curband]; + skb = qdisc->ops->peek(qdisc); + if (skb) + return skb; + } + } + return NULL; + +} + +static void +multiq_reset(struct Qdisc *sch) +{ + u16 band; + struct multiq_sched_data *q = qdisc_priv(sch); + + for (band = 0; band < q->bands; band++) + qdisc_reset(q->queues[band]); + sch->q.qlen = 0; + q->curband = 0; +} + +static void +multiq_destroy(struct Qdisc *sch) +{ + int band; + struct multiq_sched_data *q = qdisc_priv(sch); + + tcf_block_put(q->block); + for (band = 0; band < q->bands; band++) + qdisc_put(q->queues[band]); + + kfree(q->queues); +} + +static int multiq_tune(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + struct tc_multiq_qopt *qopt; + int i; + + if (!netif_is_multiqueue(qdisc_dev(sch))) + return -EOPNOTSUPP; + if (nla_len(opt) < sizeof(*qopt)) + return -EINVAL; + + qopt = nla_data(opt); + + qopt->bands = qdisc_dev(sch)->real_num_tx_queues; + + sch_tree_lock(sch); + q->bands = qopt->bands; + for (i = q->bands; i < q->max_bands; i++) { + if (q->queues[i] != &noop_qdisc) { + struct Qdisc *child = q->queues[i]; + q->queues[i] = &noop_qdisc; + qdisc_tree_reduce_backlog(child, child->q.qlen, + child->qstats.backlog); + qdisc_put(child); + } + } + + sch_tree_unlock(sch); + + for (i = 0; i < q->bands; i++) { + if (q->queues[i] == &noop_qdisc) { + struct Qdisc *child, *old; + child = qdisc_create_dflt(sch->dev_queue, + &pfifo_qdisc_ops, + TC_H_MAKE(sch->handle, + i + 1), extack); + if (child) { + sch_tree_lock(sch); + old = q->queues[i]; + q->queues[i] = child; + if (child != &noop_qdisc) + qdisc_hash_add(child, true); + + if (old != &noop_qdisc) { + qdisc_tree_reduce_backlog(old, + old->q.qlen, + old->qstats.backlog); + qdisc_put(old); + } + sch_tree_unlock(sch); + } + } + } + return 0; +} + +static int multiq_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + int i, err; + + q->queues = NULL; + + if (!opt) + return -EINVAL; + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + q->max_bands = qdisc_dev(sch)->num_tx_queues; + + q->queues = kcalloc(q->max_bands, sizeof(struct Qdisc *), GFP_KERNEL); + if (!q->queues) + return -ENOBUFS; + for (i = 0; i < q->max_bands; i++) + q->queues[i] = &noop_qdisc; + + return multiq_tune(sch, opt, extack); +} + +static int multiq_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + unsigned char *b = skb_tail_pointer(skb); + struct tc_multiq_qopt opt; + + opt.bands = q->bands; + opt.max_bands = q->max_bands; + + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int multiq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + unsigned long band = arg - 1; + + if (new == NULL) + new = &noop_qdisc; + + *old = qdisc_replace(sch, new, &q->queues[band]); + return 0; +} + +static struct Qdisc * +multiq_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + unsigned long band = arg - 1; + + return q->queues[band]; +} + +static unsigned long multiq_find(struct Qdisc *sch, u32 classid) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + unsigned long band = TC_H_MIN(classid); + + if (band - 1 >= q->bands) + return 0; + return band; +} + +static unsigned long multiq_bind(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + return multiq_find(sch, classid); +} + + +static void multiq_unbind(struct Qdisc *q, unsigned long cl) +{ +} + +static int multiq_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + + tcm->tcm_handle |= TC_H_MIN(cl); + tcm->tcm_info = q->queues[cl - 1]->handle; + return 0; +} + +static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + struct Qdisc *cl_q; + + cl_q = q->queues[cl - 1]; + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, cl_q->cpu_bstats, &cl_q->bstats) < 0 || + gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0) + return -1; + + return 0; +} + +static void multiq_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + int band; + + if (arg->stop) + return; + + for (band = 0; band < q->bands; band++) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, band + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } +} + +static struct tcf_block *multiq_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct multiq_sched_data *q = qdisc_priv(sch); + + if (cl) + return NULL; + return q->block; +} + +static const struct Qdisc_class_ops multiq_class_ops = { + .graft = multiq_graft, + .leaf = multiq_leaf, + .find = multiq_find, + .walk = multiq_walk, + .tcf_block = multiq_tcf_block, + .bind_tcf = multiq_bind, + .unbind_tcf = multiq_unbind, + .dump = multiq_dump_class, + .dump_stats = multiq_dump_class_stats, +}; + +static struct Qdisc_ops multiq_qdisc_ops __read_mostly = { + .next = NULL, + .cl_ops = &multiq_class_ops, + .id = "multiq", + .priv_size = sizeof(struct multiq_sched_data), + .enqueue = multiq_enqueue, + .dequeue = multiq_dequeue, + .peek = multiq_peek, + .init = multiq_init, + .reset = multiq_reset, + .destroy = multiq_destroy, + .change = multiq_tune, + .dump = multiq_dump, + .owner = THIS_MODULE, +}; + +static int __init multiq_module_init(void) +{ + return register_qdisc(&multiq_qdisc_ops); +} + +static void __exit multiq_module_exit(void) +{ + unregister_qdisc(&multiq_qdisc_ops); +} + +module_init(multiq_module_init) +module_exit(multiq_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c new file mode 100644 index 000000000..ad400f4f9 --- /dev/null +++ b/net/sched/sch_netem.c @@ -0,0 +1,1275 @@ +/* + * net/sched/sch_netem.c Network emulator + * + * 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. + * + * Many of the algorithms and ideas for this came from + * NIST Net which is not copyrighted. + * + * Authors: Stephen Hemminger <shemminger@osdl.org> + * Catalin(ux aka Dino) BOIE <catab at umbrella dot ro> + */ + +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/vmalloc.h> +#include <linux/rtnetlink.h> +#include <linux/reciprocal_div.h> +#include <linux/rbtree.h> + +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/inet_ecn.h> + +#define VERSION "1.3" + +/* Network Emulation Queuing algorithm. + ==================================== + + Sources: [1] Mark Carson, Darrin Santay, "NIST Net - A Linux-based + Network Emulation Tool + [2] Luigi Rizzo, DummyNet for FreeBSD + + ---------------------------------------------------------------- + + This started out as a simple way to delay outgoing packets to + test TCP but has grown to include most of the functionality + of a full blown network emulator like NISTnet. It can delay + packets and add random jitter (and correlation). The random + distribution can be loaded from a table as well to provide + normal, Pareto, or experimental curves. Packet loss, + duplication, and reordering can also be emulated. + + This qdisc does not do classification that can be handled in + layering other disciplines. It does not need to do bandwidth + control either since that can be handled by using token + bucket or other rate control. + + Correlated Loss Generator models + + Added generation of correlated loss according to the + "Gilbert-Elliot" model, a 4-state markov model. + + References: + [1] NetemCLG Home http://netgroup.uniroma2.it/NetemCLG + [2] S. Salsano, F. Ludovici, A. Ordine, "Definition of a general + and intuitive loss model for packet networks and its implementation + in the Netem module in the Linux kernel", available in [1] + + Authors: Stefano Salsano <stefano.salsano at uniroma2.it + Fabio Ludovici <fabio.ludovici at yahoo.it> +*/ + +struct disttable { + u32 size; + s16 table[0]; +}; + +struct netem_sched_data { + /* internal t(ime)fifo qdisc uses t_root and sch->limit */ + struct rb_root t_root; + + /* optional qdisc for classful handling (NULL at netem init) */ + struct Qdisc *qdisc; + + struct qdisc_watchdog watchdog; + + s64 latency; + s64 jitter; + + u32 loss; + u32 ecn; + u32 limit; + u32 counter; + u32 gap; + u32 duplicate; + u32 reorder; + u32 corrupt; + u64 rate; + s32 packet_overhead; + u32 cell_size; + struct reciprocal_value cell_size_reciprocal; + s32 cell_overhead; + + struct crndstate { + u32 last; + u32 rho; + } delay_cor, loss_cor, dup_cor, reorder_cor, corrupt_cor; + + struct disttable *delay_dist; + + enum { + CLG_RANDOM, + CLG_4_STATES, + CLG_GILB_ELL, + } loss_model; + + enum { + TX_IN_GAP_PERIOD = 1, + TX_IN_BURST_PERIOD, + LOST_IN_GAP_PERIOD, + LOST_IN_BURST_PERIOD, + } _4_state_model; + + enum { + GOOD_STATE = 1, + BAD_STATE, + } GE_state_model; + + /* Correlated Loss Generation models */ + struct clgstate { + /* state of the Markov chain */ + u8 state; + + /* 4-states and Gilbert-Elliot models */ + u32 a1; /* p13 for 4-states or p for GE */ + u32 a2; /* p31 for 4-states or r for GE */ + u32 a3; /* p32 for 4-states or h for GE */ + u32 a4; /* p14 for 4-states or 1-k for GE */ + u32 a5; /* p23 used only in 4-states */ + } clg; + + struct tc_netem_slot slot_config; + struct slotstate { + u64 slot_next; + s32 packets_left; + s32 bytes_left; + } slot; + + struct disttable *slot_dist; +}; + +/* Time stamp put into socket buffer control block + * Only valid when skbs are in our internal t(ime)fifo queue. + * + * As skb->rbnode uses same storage than skb->next, skb->prev and skb->tstamp, + * and skb->next & skb->prev are scratch space for a qdisc, + * we save skb->tstamp value in skb->cb[] before destroying it. + */ +struct netem_skb_cb { + u64 time_to_send; +}; + +static inline struct netem_skb_cb *netem_skb_cb(struct sk_buff *skb) +{ + /* we assume we can use skb next/prev/tstamp as storage for rb_node */ + qdisc_cb_private_validate(skb, sizeof(struct netem_skb_cb)); + return (struct netem_skb_cb *)qdisc_skb_cb(skb)->data; +} + +/* init_crandom - initialize correlated random number generator + * Use entropy source for initial seed. + */ +static void init_crandom(struct crndstate *state, unsigned long rho) +{ + state->rho = rho; + state->last = prandom_u32(); +} + +/* get_crandom - correlated random number generator + * Next number depends on last value. + * rho is scaled to avoid floating point. + */ +static u32 get_crandom(struct crndstate *state) +{ + u64 value, rho; + unsigned long answer; + + if (!state || state->rho == 0) /* no correlation */ + return prandom_u32(); + + value = prandom_u32(); + rho = (u64)state->rho + 1; + answer = (value * ((1ull<<32) - rho) + state->last * rho) >> 32; + state->last = answer; + return answer; +} + +/* loss_4state - 4-state model loss generator + * Generates losses according to the 4-state Markov chain adopted in + * the GI (General and Intuitive) loss model. + */ +static bool loss_4state(struct netem_sched_data *q) +{ + struct clgstate *clg = &q->clg; + u32 rnd = prandom_u32(); + + /* + * Makes a comparison between rnd and the transition + * probabilities outgoing from the current state, then decides the + * next state and if the next packet has to be transmitted or lost. + * The four states correspond to: + * TX_IN_GAP_PERIOD => successfully transmitted packets within a gap period + * LOST_IN_BURST_PERIOD => isolated losses within a gap period + * LOST_IN_GAP_PERIOD => lost packets within a burst period + * TX_IN_GAP_PERIOD => successfully transmitted packets within a burst period + */ + switch (clg->state) { + case TX_IN_GAP_PERIOD: + if (rnd < clg->a4) { + clg->state = LOST_IN_BURST_PERIOD; + return true; + } else if (clg->a4 < rnd && rnd < clg->a1 + clg->a4) { + clg->state = LOST_IN_GAP_PERIOD; + return true; + } else if (clg->a1 + clg->a4 < rnd) { + clg->state = TX_IN_GAP_PERIOD; + } + + break; + case TX_IN_BURST_PERIOD: + if (rnd < clg->a5) { + clg->state = LOST_IN_GAP_PERIOD; + return true; + } else { + clg->state = TX_IN_BURST_PERIOD; + } + + break; + case LOST_IN_GAP_PERIOD: + if (rnd < clg->a3) + clg->state = TX_IN_BURST_PERIOD; + else if (clg->a3 < rnd && rnd < clg->a2 + clg->a3) { + clg->state = TX_IN_GAP_PERIOD; + } else if (clg->a2 + clg->a3 < rnd) { + clg->state = LOST_IN_GAP_PERIOD; + return true; + } + break; + case LOST_IN_BURST_PERIOD: + clg->state = TX_IN_GAP_PERIOD; + break; + } + + return false; +} + +/* loss_gilb_ell - Gilbert-Elliot model loss generator + * Generates losses according to the Gilbert-Elliot loss model or + * its special cases (Gilbert or Simple Gilbert) + * + * Makes a comparison between random number and the transition + * probabilities outgoing from the current state, then decides the + * next state. A second random number is extracted and the comparison + * with the loss probability of the current state decides if the next + * packet will be transmitted or lost. + */ +static bool loss_gilb_ell(struct netem_sched_data *q) +{ + struct clgstate *clg = &q->clg; + + switch (clg->state) { + case GOOD_STATE: + if (prandom_u32() < clg->a1) + clg->state = BAD_STATE; + if (prandom_u32() < clg->a4) + return true; + break; + case BAD_STATE: + if (prandom_u32() < clg->a2) + clg->state = GOOD_STATE; + if (prandom_u32() > clg->a3) + return true; + } + + return false; +} + +static bool loss_event(struct netem_sched_data *q) +{ + switch (q->loss_model) { + case CLG_RANDOM: + /* Random packet drop 0 => none, ~0 => all */ + return q->loss && q->loss >= get_crandom(&q->loss_cor); + + case CLG_4_STATES: + /* 4state loss model algorithm (used also for GI model) + * Extracts a value from the markov 4 state loss generator, + * if it is 1 drops a packet and if needed writes the event in + * the kernel logs + */ + return loss_4state(q); + + case CLG_GILB_ELL: + /* Gilbert-Elliot loss model algorithm + * Extracts a value from the Gilbert-Elliot loss generator, + * if it is 1 drops a packet and if needed writes the event in + * the kernel logs + */ + return loss_gilb_ell(q); + } + + return false; /* not reached */ +} + + +/* tabledist - return a pseudo-randomly distributed value with mean mu and + * std deviation sigma. Uses table lookup to approximate the desired + * distribution, and a uniformly-distributed pseudo-random source. + */ +static s64 tabledist(s64 mu, s32 sigma, + struct crndstate *state, + const struct disttable *dist) +{ + s64 x; + long t; + u32 rnd; + + if (sigma == 0) + return mu; + + rnd = get_crandom(state); + + /* default uniform distribution */ + if (dist == NULL) + return ((rnd % (2 * (u32)sigma)) + mu) - sigma; + + t = dist->table[rnd % dist->size]; + x = (sigma % NETEM_DIST_SCALE) * t; + if (x >= 0) + x += NETEM_DIST_SCALE/2; + else + x -= NETEM_DIST_SCALE/2; + + return x / NETEM_DIST_SCALE + (sigma / NETEM_DIST_SCALE) * t + mu; +} + +static u64 packet_time_ns(u64 len, const struct netem_sched_data *q) +{ + len += q->packet_overhead; + + if (q->cell_size) { + u32 cells = reciprocal_divide(len, q->cell_size_reciprocal); + + if (len > cells * q->cell_size) /* extra cell needed for remainder */ + cells++; + len = cells * (q->cell_size + q->cell_overhead); + } + + return div64_u64(len * NSEC_PER_SEC, q->rate); +} + +static void tfifo_reset(struct Qdisc *sch) +{ + struct netem_sched_data *q = qdisc_priv(sch); + struct rb_node *p = rb_first(&q->t_root); + + while (p) { + struct sk_buff *skb = rb_to_skb(p); + + p = rb_next(p); + rb_erase(&skb->rbnode, &q->t_root); + rtnl_kfree_skbs(skb, skb); + } +} + +static void tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch) +{ + struct netem_sched_data *q = qdisc_priv(sch); + u64 tnext = netem_skb_cb(nskb)->time_to_send; + struct rb_node **p = &q->t_root.rb_node, *parent = NULL; + + while (*p) { + struct sk_buff *skb; + + parent = *p; + skb = rb_to_skb(parent); + if (tnext >= netem_skb_cb(skb)->time_to_send) + p = &parent->rb_right; + else + p = &parent->rb_left; + } + rb_link_node(&nskb->rbnode, parent, p); + rb_insert_color(&nskb->rbnode, &q->t_root); + sch->q.qlen++; +} + +/* netem can't properly corrupt a megapacket (like we get from GSO), so instead + * when we statistically choose to corrupt one, we instead segment it, returning + * the first packet to be corrupted, and re-enqueue the remaining frames + */ +static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct sk_buff *segs; + netdev_features_t features = netif_skb_features(skb); + + segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); + + if (IS_ERR_OR_NULL(segs)) { + qdisc_drop(skb, sch, to_free); + return NULL; + } + consume_skb(skb); + return segs; +} + +static void netem_enqueue_skb_head(struct qdisc_skb_head *qh, struct sk_buff *skb) +{ + skb->next = qh->head; + + if (!qh->head) + qh->tail = skb; + qh->head = skb; + qh->qlen++; +} + +/* + * Insert one skb into qdisc. + * Note: parent depends on return value to account for queue length. + * NET_XMIT_DROP: queue length didn't change. + * NET_XMIT_SUCCESS: one skb was queued. + */ +static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct netem_sched_data *q = qdisc_priv(sch); + /* We don't fill cb now as skb_unshare() may invalidate it */ + struct netem_skb_cb *cb; + struct sk_buff *skb2; + struct sk_buff *segs = NULL; + unsigned int prev_len = qdisc_pkt_len(skb); + int count = 1; + int rc = NET_XMIT_SUCCESS; + int rc_drop = NET_XMIT_DROP; + + /* Do not fool qdisc_drop_all() */ + skb->prev = NULL; + + /* Random duplication */ + if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor)) + ++count; + + /* Drop packet? */ + if (loss_event(q)) { + if (q->ecn && INET_ECN_set_ce(skb)) + qdisc_qstats_drop(sch); /* mark packet */ + else + --count; + } + if (count == 0) { + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + } + + /* If a delay is expected, orphan the skb. (orphaning usually takes + * place at TX completion time, so _before_ the link transit delay) + */ + if (q->latency || q->jitter || q->rate) + skb_orphan_partial(skb); + + /* + * If we need to duplicate packet, then re-insert at top of the + * qdisc tree, since parent queuer expects that only one + * skb will be queued. + */ + if (count > 1 && (skb2 = skb_clone(skb, GFP_ATOMIC)) != NULL) { + struct Qdisc *rootq = qdisc_root_bh(sch); + u32 dupsave = q->duplicate; /* prevent duplicating a dup... */ + + q->duplicate = 0; + rootq->enqueue(skb2, rootq, to_free); + q->duplicate = dupsave; + rc_drop = NET_XMIT_SUCCESS; + } + + /* + * Randomized packet corruption. + * Make copy if needed since we are modifying + * If packet is going to be hardware checksummed, then + * do it now in software before we mangle it. + */ + if (q->corrupt && q->corrupt >= get_crandom(&q->corrupt_cor)) { + if (skb_is_gso(skb)) { + segs = netem_segment(skb, sch, to_free); + if (!segs) + return rc_drop; + qdisc_skb_cb(segs)->pkt_len = segs->len; + } else { + segs = skb; + } + + skb = segs; + segs = segs->next; + + skb = skb_unshare(skb, GFP_ATOMIC); + if (unlikely(!skb)) { + qdisc_qstats_drop(sch); + goto finish_segs; + } + if (skb->ip_summed == CHECKSUM_PARTIAL && + skb_checksum_help(skb)) { + qdisc_drop(skb, sch, to_free); + skb = NULL; + goto finish_segs; + } + + skb->data[prandom_u32() % skb_headlen(skb)] ^= + 1<<(prandom_u32() % 8); + } + + if (unlikely(sch->q.qlen >= sch->limit)) { + qdisc_drop_all(skb, sch, to_free); + return rc_drop; + } + + qdisc_qstats_backlog_inc(sch, skb); + + cb = netem_skb_cb(skb); + if (q->gap == 0 || /* not doing reordering */ + q->counter < q->gap - 1 || /* inside last reordering gap */ + q->reorder < get_crandom(&q->reorder_cor)) { + u64 now; + s64 delay; + + delay = tabledist(q->latency, q->jitter, + &q->delay_cor, q->delay_dist); + + now = ktime_get_ns(); + + if (q->rate) { + struct netem_skb_cb *last = NULL; + + if (sch->q.tail) + last = netem_skb_cb(sch->q.tail); + if (q->t_root.rb_node) { + struct sk_buff *t_skb; + struct netem_skb_cb *t_last; + + t_skb = skb_rb_last(&q->t_root); + t_last = netem_skb_cb(t_skb); + if (!last || + t_last->time_to_send > last->time_to_send) { + last = t_last; + } + } + + if (last) { + /* + * Last packet in queue is reference point (now), + * calculate this time bonus and subtract + * from delay. + */ + delay -= last->time_to_send - now; + delay = max_t(s64, 0, delay); + now = last->time_to_send; + } + + delay += packet_time_ns(qdisc_pkt_len(skb), q); + } + + cb->time_to_send = now + delay; + ++q->counter; + tfifo_enqueue(skb, sch); + } else { + /* + * Do re-ordering by putting one out of N packets at the front + * of the queue. + */ + cb->time_to_send = ktime_get_ns(); + q->counter = 0; + + netem_enqueue_skb_head(&sch->q, skb); + sch->qstats.requeues++; + } + +finish_segs: + if (segs) { + unsigned int len, last_len; + int nb; + + len = skb ? skb->len : 0; + nb = skb ? 1 : 0; + + while (segs) { + skb2 = segs->next; + segs->next = NULL; + qdisc_skb_cb(segs)->pkt_len = segs->len; + last_len = segs->len; + rc = qdisc_enqueue(segs, sch, to_free); + if (rc != NET_XMIT_SUCCESS) { + if (net_xmit_drop_count(rc)) + qdisc_qstats_drop(sch); + } else { + nb++; + len += last_len; + } + segs = skb2; + } + /* Parent qdiscs accounted for 1 skb of size @prev_len */ + qdisc_tree_reduce_backlog(sch, -(nb - 1), -(len - prev_len)); + } else if (!skb) { + return NET_XMIT_DROP; + } + return NET_XMIT_SUCCESS; +} + +/* Delay the next round with a new future slot with a + * correct number of bytes and packets. + */ + +static void get_slot_next(struct netem_sched_data *q, u64 now) +{ + s64 next_delay; + + if (!q->slot_dist) + next_delay = q->slot_config.min_delay + + (prandom_u32() * + (q->slot_config.max_delay - + q->slot_config.min_delay) >> 32); + else + next_delay = tabledist(q->slot_config.dist_delay, + (s32)(q->slot_config.dist_jitter), + NULL, q->slot_dist); + + q->slot.slot_next = now + next_delay; + q->slot.packets_left = q->slot_config.max_packets; + q->slot.bytes_left = q->slot_config.max_bytes; +} + +static struct sk_buff *netem_dequeue(struct Qdisc *sch) +{ + struct netem_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + struct rb_node *p; + +tfifo_dequeue: + skb = __qdisc_dequeue_head(&sch->q); + if (skb) { + qdisc_qstats_backlog_dec(sch, skb); +deliver: + qdisc_bstats_update(sch, skb); + return skb; + } + p = rb_first(&q->t_root); + if (p) { + u64 time_to_send; + u64 now = ktime_get_ns(); + + skb = rb_to_skb(p); + + /* if more time remaining? */ + time_to_send = netem_skb_cb(skb)->time_to_send; + if (q->slot.slot_next && q->slot.slot_next < time_to_send) + get_slot_next(q, now); + + if (time_to_send <= now && q->slot.slot_next <= now) { + rb_erase(p, &q->t_root); + sch->q.qlen--; + qdisc_qstats_backlog_dec(sch, skb); + skb->next = NULL; + skb->prev = NULL; + /* skb->dev shares skb->rbnode area, + * we need to restore its value. + */ + skb->dev = qdisc_dev(sch); + +#ifdef CONFIG_NET_CLS_ACT + /* + * If it's at ingress let's pretend the delay is + * from the network (tstamp will be updated). + */ + if (skb->tc_redirected && skb->tc_from_ingress) + skb->tstamp = 0; +#endif + + if (q->slot.slot_next) { + q->slot.packets_left--; + q->slot.bytes_left -= qdisc_pkt_len(skb); + if (q->slot.packets_left <= 0 || + q->slot.bytes_left <= 0) + get_slot_next(q, now); + } + + if (q->qdisc) { + unsigned int pkt_len = qdisc_pkt_len(skb); + struct sk_buff *to_free = NULL; + int err; + + err = qdisc_enqueue(skb, q->qdisc, &to_free); + kfree_skb_list(to_free); + if (err != NET_XMIT_SUCCESS && + net_xmit_drop_count(err)) { + qdisc_qstats_drop(sch); + qdisc_tree_reduce_backlog(sch, 1, + pkt_len); + } + goto tfifo_dequeue; + } + goto deliver; + } + + if (q->qdisc) { + skb = q->qdisc->ops->dequeue(q->qdisc); + if (skb) + goto deliver; + } + + qdisc_watchdog_schedule_ns(&q->watchdog, + max(time_to_send, + q->slot.slot_next)); + } + + if (q->qdisc) { + skb = q->qdisc->ops->dequeue(q->qdisc); + if (skb) + goto deliver; + } + return NULL; +} + +static void netem_reset(struct Qdisc *sch) +{ + struct netem_sched_data *q = qdisc_priv(sch); + + qdisc_reset_queue(sch); + tfifo_reset(sch); + if (q->qdisc) + qdisc_reset(q->qdisc); + qdisc_watchdog_cancel(&q->watchdog); +} + +static void dist_free(struct disttable *d) +{ + kvfree(d); +} + +/* + * Distribution data is a variable size payload containing + * signed 16 bit values. + */ + +static int get_dist_table(struct Qdisc *sch, struct disttable **tbl, + const struct nlattr *attr) +{ + size_t n = nla_len(attr)/sizeof(__s16); + const __s16 *data = nla_data(attr); + spinlock_t *root_lock; + struct disttable *d; + int i; + + if (!n || n > NETEM_DIST_MAX) + return -EINVAL; + + d = kvmalloc(sizeof(struct disttable) + n * sizeof(s16), GFP_KERNEL); + if (!d) + return -ENOMEM; + + d->size = n; + for (i = 0; i < n; i++) + d->table[i] = data[i]; + + root_lock = qdisc_root_sleeping_lock(sch); + + spin_lock_bh(root_lock); + swap(*tbl, d); + spin_unlock_bh(root_lock); + + dist_free(d); + return 0; +} + +static void get_slot(struct netem_sched_data *q, const struct nlattr *attr) +{ + const struct tc_netem_slot *c = nla_data(attr); + + q->slot_config = *c; + if (q->slot_config.max_packets == 0) + q->slot_config.max_packets = INT_MAX; + if (q->slot_config.max_bytes == 0) + q->slot_config.max_bytes = INT_MAX; + + /* capping dist_jitter to the range acceptable by tabledist() */ + q->slot_config.dist_jitter = min_t(__s64, INT_MAX, abs(q->slot_config.dist_jitter)); + + q->slot.packets_left = q->slot_config.max_packets; + q->slot.bytes_left = q->slot_config.max_bytes; + if (q->slot_config.min_delay | q->slot_config.max_delay | + q->slot_config.dist_jitter) + q->slot.slot_next = ktime_get_ns(); + else + q->slot.slot_next = 0; +} + +static void get_correlation(struct netem_sched_data *q, const struct nlattr *attr) +{ + const struct tc_netem_corr *c = nla_data(attr); + + init_crandom(&q->delay_cor, c->delay_corr); + init_crandom(&q->loss_cor, c->loss_corr); + init_crandom(&q->dup_cor, c->dup_corr); +} + +static void get_reorder(struct netem_sched_data *q, const struct nlattr *attr) +{ + const struct tc_netem_reorder *r = nla_data(attr); + + q->reorder = r->probability; + init_crandom(&q->reorder_cor, r->correlation); +} + +static void get_corrupt(struct netem_sched_data *q, const struct nlattr *attr) +{ + const struct tc_netem_corrupt *r = nla_data(attr); + + q->corrupt = r->probability; + init_crandom(&q->corrupt_cor, r->correlation); +} + +static void get_rate(struct netem_sched_data *q, const struct nlattr *attr) +{ + const struct tc_netem_rate *r = nla_data(attr); + + q->rate = r->rate; + q->packet_overhead = r->packet_overhead; + q->cell_size = r->cell_size; + q->cell_overhead = r->cell_overhead; + if (q->cell_size) + q->cell_size_reciprocal = reciprocal_value(q->cell_size); + else + q->cell_size_reciprocal = (struct reciprocal_value) { 0 }; +} + +static int get_loss_clg(struct netem_sched_data *q, const struct nlattr *attr) +{ + const struct nlattr *la; + int rem; + + nla_for_each_nested(la, attr, rem) { + u16 type = nla_type(la); + + switch (type) { + case NETEM_LOSS_GI: { + const struct tc_netem_gimodel *gi = nla_data(la); + + if (nla_len(la) < sizeof(struct tc_netem_gimodel)) { + pr_info("netem: incorrect gi model size\n"); + return -EINVAL; + } + + q->loss_model = CLG_4_STATES; + + q->clg.state = TX_IN_GAP_PERIOD; + q->clg.a1 = gi->p13; + q->clg.a2 = gi->p31; + q->clg.a3 = gi->p32; + q->clg.a4 = gi->p14; + q->clg.a5 = gi->p23; + break; + } + + case NETEM_LOSS_GE: { + const struct tc_netem_gemodel *ge = nla_data(la); + + if (nla_len(la) < sizeof(struct tc_netem_gemodel)) { + pr_info("netem: incorrect ge model size\n"); + return -EINVAL; + } + + q->loss_model = CLG_GILB_ELL; + q->clg.state = GOOD_STATE; + q->clg.a1 = ge->p; + q->clg.a2 = ge->r; + q->clg.a3 = ge->h; + q->clg.a4 = ge->k1; + break; + } + + default: + pr_info("netem: unknown loss type %u\n", type); + return -EINVAL; + } + } + + return 0; +} + +static const struct nla_policy netem_policy[TCA_NETEM_MAX + 1] = { + [TCA_NETEM_CORR] = { .len = sizeof(struct tc_netem_corr) }, + [TCA_NETEM_REORDER] = { .len = sizeof(struct tc_netem_reorder) }, + [TCA_NETEM_CORRUPT] = { .len = sizeof(struct tc_netem_corrupt) }, + [TCA_NETEM_RATE] = { .len = sizeof(struct tc_netem_rate) }, + [TCA_NETEM_LOSS] = { .type = NLA_NESTED }, + [TCA_NETEM_ECN] = { .type = NLA_U32 }, + [TCA_NETEM_RATE64] = { .type = NLA_U64 }, + [TCA_NETEM_LATENCY64] = { .type = NLA_S64 }, + [TCA_NETEM_JITTER64] = { .type = NLA_S64 }, + [TCA_NETEM_SLOT] = { .len = sizeof(struct tc_netem_slot) }, +}; + +static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla, + const struct nla_policy *policy, int len) +{ + int nested_len = nla_len(nla) - NLA_ALIGN(len); + + if (nested_len < 0) { + pr_info("netem: invalid attributes len %d\n", nested_len); + return -EINVAL; + } + + if (nested_len >= nla_attr_size(0)) + return nla_parse(tb, maxtype, nla_data(nla) + NLA_ALIGN(len), + nested_len, policy, NULL); + + memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1)); + return 0; +} + +/* Parse netlink message to set options */ +static int netem_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct netem_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_NETEM_MAX + 1]; + struct tc_netem_qopt *qopt; + struct clgstate old_clg; + int old_loss_model = CLG_RANDOM; + int ret; + + if (opt == NULL) + return -EINVAL; + + qopt = nla_data(opt); + ret = parse_attr(tb, TCA_NETEM_MAX, opt, netem_policy, sizeof(*qopt)); + if (ret < 0) + return ret; + + /* backup q->clg and q->loss_model */ + old_clg = q->clg; + old_loss_model = q->loss_model; + + if (tb[TCA_NETEM_LOSS]) { + ret = get_loss_clg(q, tb[TCA_NETEM_LOSS]); + if (ret) { + q->loss_model = old_loss_model; + return ret; + } + } else { + q->loss_model = CLG_RANDOM; + } + + if (tb[TCA_NETEM_DELAY_DIST]) { + ret = get_dist_table(sch, &q->delay_dist, + tb[TCA_NETEM_DELAY_DIST]); + if (ret) + goto get_table_failure; + } + + if (tb[TCA_NETEM_SLOT_DIST]) { + ret = get_dist_table(sch, &q->slot_dist, + tb[TCA_NETEM_SLOT_DIST]); + if (ret) + goto get_table_failure; + } + + sch->limit = qopt->limit; + + q->latency = PSCHED_TICKS2NS(qopt->latency); + q->jitter = PSCHED_TICKS2NS(qopt->jitter); + q->limit = qopt->limit; + q->gap = qopt->gap; + q->counter = 0; + q->loss = qopt->loss; + q->duplicate = qopt->duplicate; + + /* for compatibility with earlier versions. + * if gap is set, need to assume 100% probability + */ + if (q->gap) + q->reorder = ~0; + + if (tb[TCA_NETEM_CORR]) + get_correlation(q, tb[TCA_NETEM_CORR]); + + if (tb[TCA_NETEM_REORDER]) + get_reorder(q, tb[TCA_NETEM_REORDER]); + + if (tb[TCA_NETEM_CORRUPT]) + get_corrupt(q, tb[TCA_NETEM_CORRUPT]); + + if (tb[TCA_NETEM_RATE]) + get_rate(q, tb[TCA_NETEM_RATE]); + + if (tb[TCA_NETEM_RATE64]) + q->rate = max_t(u64, q->rate, + nla_get_u64(tb[TCA_NETEM_RATE64])); + + if (tb[TCA_NETEM_LATENCY64]) + q->latency = nla_get_s64(tb[TCA_NETEM_LATENCY64]); + + if (tb[TCA_NETEM_JITTER64]) + q->jitter = nla_get_s64(tb[TCA_NETEM_JITTER64]); + + if (tb[TCA_NETEM_ECN]) + q->ecn = nla_get_u32(tb[TCA_NETEM_ECN]); + + if (tb[TCA_NETEM_SLOT]) + get_slot(q, tb[TCA_NETEM_SLOT]); + + /* capping jitter to the range acceptable by tabledist() */ + q->jitter = min_t(s64, abs(q->jitter), INT_MAX); + + return ret; + +get_table_failure: + /* recover clg and loss_model, in case of + * q->clg and q->loss_model were modified + * in get_loss_clg() + */ + q->clg = old_clg; + q->loss_model = old_loss_model; + return ret; +} + +static int netem_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct netem_sched_data *q = qdisc_priv(sch); + int ret; + + qdisc_watchdog_init(&q->watchdog, sch); + + if (!opt) + return -EINVAL; + + q->loss_model = CLG_RANDOM; + ret = netem_change(sch, opt, extack); + if (ret) + pr_info("netem: change failed\n"); + return ret; +} + +static void netem_destroy(struct Qdisc *sch) +{ + struct netem_sched_data *q = qdisc_priv(sch); + + qdisc_watchdog_cancel(&q->watchdog); + if (q->qdisc) + qdisc_put(q->qdisc); + dist_free(q->delay_dist); + dist_free(q->slot_dist); +} + +static int dump_loss_model(const struct netem_sched_data *q, + struct sk_buff *skb) +{ + struct nlattr *nest; + + nest = nla_nest_start(skb, TCA_NETEM_LOSS); + if (nest == NULL) + goto nla_put_failure; + + switch (q->loss_model) { + case CLG_RANDOM: + /* legacy loss model */ + nla_nest_cancel(skb, nest); + return 0; /* no data */ + + case CLG_4_STATES: { + struct tc_netem_gimodel gi = { + .p13 = q->clg.a1, + .p31 = q->clg.a2, + .p32 = q->clg.a3, + .p14 = q->clg.a4, + .p23 = q->clg.a5, + }; + + if (nla_put(skb, NETEM_LOSS_GI, sizeof(gi), &gi)) + goto nla_put_failure; + break; + } + case CLG_GILB_ELL: { + struct tc_netem_gemodel ge = { + .p = q->clg.a1, + .r = q->clg.a2, + .h = q->clg.a3, + .k1 = q->clg.a4, + }; + + if (nla_put(skb, NETEM_LOSS_GE, sizeof(ge), &ge)) + goto nla_put_failure; + break; + } + } + + nla_nest_end(skb, nest); + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int netem_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + const struct netem_sched_data *q = qdisc_priv(sch); + struct nlattr *nla = (struct nlattr *) skb_tail_pointer(skb); + struct tc_netem_qopt qopt; + struct tc_netem_corr cor; + struct tc_netem_reorder reorder; + struct tc_netem_corrupt corrupt; + struct tc_netem_rate rate; + struct tc_netem_slot slot; + + qopt.latency = min_t(psched_tdiff_t, PSCHED_NS2TICKS(q->latency), + UINT_MAX); + qopt.jitter = min_t(psched_tdiff_t, PSCHED_NS2TICKS(q->jitter), + UINT_MAX); + qopt.limit = q->limit; + qopt.loss = q->loss; + qopt.gap = q->gap; + qopt.duplicate = q->duplicate; + if (nla_put(skb, TCA_OPTIONS, sizeof(qopt), &qopt)) + goto nla_put_failure; + + if (nla_put(skb, TCA_NETEM_LATENCY64, sizeof(q->latency), &q->latency)) + goto nla_put_failure; + + if (nla_put(skb, TCA_NETEM_JITTER64, sizeof(q->jitter), &q->jitter)) + goto nla_put_failure; + + cor.delay_corr = q->delay_cor.rho; + cor.loss_corr = q->loss_cor.rho; + cor.dup_corr = q->dup_cor.rho; + if (nla_put(skb, TCA_NETEM_CORR, sizeof(cor), &cor)) + goto nla_put_failure; + + reorder.probability = q->reorder; + reorder.correlation = q->reorder_cor.rho; + if (nla_put(skb, TCA_NETEM_REORDER, sizeof(reorder), &reorder)) + goto nla_put_failure; + + corrupt.probability = q->corrupt; + corrupt.correlation = q->corrupt_cor.rho; + if (nla_put(skb, TCA_NETEM_CORRUPT, sizeof(corrupt), &corrupt)) + goto nla_put_failure; + + if (q->rate >= (1ULL << 32)) { + if (nla_put_u64_64bit(skb, TCA_NETEM_RATE64, q->rate, + TCA_NETEM_PAD)) + goto nla_put_failure; + rate.rate = ~0U; + } else { + rate.rate = q->rate; + } + rate.packet_overhead = q->packet_overhead; + rate.cell_size = q->cell_size; + rate.cell_overhead = q->cell_overhead; + if (nla_put(skb, TCA_NETEM_RATE, sizeof(rate), &rate)) + goto nla_put_failure; + + if (q->ecn && nla_put_u32(skb, TCA_NETEM_ECN, q->ecn)) + goto nla_put_failure; + + if (dump_loss_model(q, skb) != 0) + goto nla_put_failure; + + if (q->slot_config.min_delay | q->slot_config.max_delay | + q->slot_config.dist_jitter) { + slot = q->slot_config; + if (slot.max_packets == INT_MAX) + slot.max_packets = 0; + if (slot.max_bytes == INT_MAX) + slot.max_bytes = 0; + if (nla_put(skb, TCA_NETEM_SLOT, sizeof(slot), &slot)) + goto nla_put_failure; + } + + return nla_nest_end(skb, nla); + +nla_put_failure: + nlmsg_trim(skb, nla); + return -1; +} + +static int netem_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct netem_sched_data *q = qdisc_priv(sch); + + if (cl != 1 || !q->qdisc) /* only one class */ + return -ENOENT; + + tcm->tcm_handle |= TC_H_MIN(1); + tcm->tcm_info = q->qdisc->handle; + + return 0; +} + +static int netem_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct netem_sched_data *q = qdisc_priv(sch); + + *old = qdisc_replace(sch, new, &q->qdisc); + return 0; +} + +static struct Qdisc *netem_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct netem_sched_data *q = qdisc_priv(sch); + return q->qdisc; +} + +static unsigned long netem_find(struct Qdisc *sch, u32 classid) +{ + return 1; +} + +static void netem_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + if (!walker->stop) { + if (walker->count >= walker->skip) + if (walker->fn(sch, 1, walker) < 0) { + walker->stop = 1; + return; + } + walker->count++; + } +} + +static const struct Qdisc_class_ops netem_class_ops = { + .graft = netem_graft, + .leaf = netem_leaf, + .find = netem_find, + .walk = netem_walk, + .dump = netem_dump_class, +}; + +static struct Qdisc_ops netem_qdisc_ops __read_mostly = { + .id = "netem", + .cl_ops = &netem_class_ops, + .priv_size = sizeof(struct netem_sched_data), + .enqueue = netem_enqueue, + .dequeue = netem_dequeue, + .peek = qdisc_peek_dequeued, + .init = netem_init, + .reset = netem_reset, + .destroy = netem_destroy, + .change = netem_change, + .dump = netem_dump, + .owner = THIS_MODULE, +}; + + +static int __init netem_module_init(void) +{ + pr_info("netem: version " VERSION "\n"); + return register_qdisc(&netem_qdisc_ops); +} +static void __exit netem_module_exit(void) +{ + unregister_qdisc(&netem_qdisc_ops); +} +module_init(netem_module_init) +module_exit(netem_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c new file mode 100644 index 000000000..18d30bb86 --- /dev/null +++ b/net/sched/sch_pie.c @@ -0,0 +1,572 @@ +/* Copyright (C) 2013 Cisco Systems, Inc, 2013. + * + * 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. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Author: Vijay Subramanian <vijaynsu@cisco.com> + * Author: Mythili Prabhu <mysuryan@cisco.com> + * + * ECN support is added by Naeem Khademi <naeemk@ifi.uio.no> + * University of Oslo, Norway. + * + * References: + * IETF draft submission: http://tools.ietf.org/html/draft-pan-aqm-pie-00 + * IEEE Conference on High Performance Switching and Routing 2013 : + * "PIE: A * Lightweight Control Scheme to Address the Bufferbloat Problem" + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> +#include <net/inet_ecn.h> + +#define QUEUE_THRESHOLD 10000 +#define DQCOUNT_INVALID -1 +#define MAX_PROB 0xffffffff +#define PIE_SCALE 8 + +/* parameters used */ +struct pie_params { + psched_time_t target; /* user specified target delay in pschedtime */ + u32 tupdate; /* timer frequency (in jiffies) */ + u32 limit; /* number of packets that can be enqueued */ + u32 alpha; /* alpha and beta are between 0 and 32 */ + u32 beta; /* and are used for shift relative to 1 */ + bool ecn; /* true if ecn is enabled */ + bool bytemode; /* to scale drop early prob based on pkt size */ +}; + +/* variables used */ +struct pie_vars { + u32 prob; /* probability but scaled by u32 limit. */ + psched_time_t burst_time; + psched_time_t qdelay; + psched_time_t qdelay_old; + u64 dq_count; /* measured in bytes */ + psched_time_t dq_tstamp; /* drain rate */ + u32 avg_dq_rate; /* bytes per pschedtime tick,scaled */ + u32 qlen_old; /* in bytes */ +}; + +/* statistics gathering */ +struct pie_stats { + u32 packets_in; /* total number of packets enqueued */ + u32 dropped; /* packets dropped due to pie_action */ + u32 overlimit; /* dropped due to lack of space in queue */ + u32 maxq; /* maximum queue size */ + u32 ecn_mark; /* packets marked with ECN */ +}; + +/* private data for the Qdisc */ +struct pie_sched_data { + struct pie_params params; + struct pie_vars vars; + struct pie_stats stats; + struct timer_list adapt_timer; + struct Qdisc *sch; +}; + +static void pie_params_init(struct pie_params *params) +{ + params->alpha = 2; + params->beta = 20; + params->tupdate = usecs_to_jiffies(30 * USEC_PER_MSEC); /* 30 ms */ + params->limit = 1000; /* default of 1000 packets */ + params->target = PSCHED_NS2TICKS(20 * NSEC_PER_MSEC); /* 20 ms */ + params->ecn = false; + params->bytemode = false; +} + +static void pie_vars_init(struct pie_vars *vars) +{ + vars->dq_count = DQCOUNT_INVALID; + vars->avg_dq_rate = 0; + /* default of 100 ms in pschedtime */ + vars->burst_time = PSCHED_NS2TICKS(100 * NSEC_PER_MSEC); +} + +static bool drop_early(struct Qdisc *sch, u32 packet_size) +{ + struct pie_sched_data *q = qdisc_priv(sch); + u32 rnd; + u32 local_prob = q->vars.prob; + u32 mtu = psched_mtu(qdisc_dev(sch)); + + /* If there is still burst allowance left skip random early drop */ + if (q->vars.burst_time > 0) + return false; + + /* If current delay is less than half of target, and + * if drop prob is low already, disable early_drop + */ + if ((q->vars.qdelay < q->params.target / 2) + && (q->vars.prob < MAX_PROB / 5)) + return false; + + /* If we have fewer than 2 mtu-sized packets, disable drop_early, + * similar to min_th in RED + */ + if (sch->qstats.backlog < 2 * mtu) + return false; + + /* If bytemode is turned on, use packet size to compute new + * probablity. Smaller packets will have lower drop prob in this case + */ + if (q->params.bytemode && packet_size <= mtu) + local_prob = (local_prob / mtu) * packet_size; + else + local_prob = q->vars.prob; + + rnd = prandom_u32(); + if (rnd < local_prob) + return true; + + return false; +} + +static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct pie_sched_data *q = qdisc_priv(sch); + bool enqueue = false; + + if (unlikely(qdisc_qlen(sch) >= sch->limit)) { + q->stats.overlimit++; + goto out; + } + + if (!drop_early(sch, skb->len)) { + enqueue = true; + } else if (q->params.ecn && (q->vars.prob <= MAX_PROB / 10) && + INET_ECN_set_ce(skb)) { + /* If packet is ecn capable, mark it if drop probability + * is lower than 10%, else drop it. + */ + q->stats.ecn_mark++; + enqueue = true; + } + + /* we can enqueue the packet */ + if (enqueue) { + q->stats.packets_in++; + if (qdisc_qlen(sch) > q->stats.maxq) + q->stats.maxq = qdisc_qlen(sch); + + return qdisc_enqueue_tail(skb, sch); + } + +out: + q->stats.dropped++; + return qdisc_drop(skb, sch, to_free); +} + +static const struct nla_policy pie_policy[TCA_PIE_MAX + 1] = { + [TCA_PIE_TARGET] = {.type = NLA_U32}, + [TCA_PIE_LIMIT] = {.type = NLA_U32}, + [TCA_PIE_TUPDATE] = {.type = NLA_U32}, + [TCA_PIE_ALPHA] = {.type = NLA_U32}, + [TCA_PIE_BETA] = {.type = NLA_U32}, + [TCA_PIE_ECN] = {.type = NLA_U32}, + [TCA_PIE_BYTEMODE] = {.type = NLA_U32}, +}; + +static int pie_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct pie_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_PIE_MAX + 1]; + unsigned int qlen, dropped = 0; + int err; + + if (!opt) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_PIE_MAX, opt, pie_policy, NULL); + if (err < 0) + return err; + + sch_tree_lock(sch); + + /* convert from microseconds to pschedtime */ + if (tb[TCA_PIE_TARGET]) { + /* target is in us */ + u32 target = nla_get_u32(tb[TCA_PIE_TARGET]); + + /* convert to pschedtime */ + q->params.target = PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC); + } + + /* tupdate is in jiffies */ + if (tb[TCA_PIE_TUPDATE]) + q->params.tupdate = usecs_to_jiffies(nla_get_u32(tb[TCA_PIE_TUPDATE])); + + if (tb[TCA_PIE_LIMIT]) { + u32 limit = nla_get_u32(tb[TCA_PIE_LIMIT]); + + q->params.limit = limit; + sch->limit = limit; + } + + if (tb[TCA_PIE_ALPHA]) + q->params.alpha = nla_get_u32(tb[TCA_PIE_ALPHA]); + + if (tb[TCA_PIE_BETA]) + q->params.beta = nla_get_u32(tb[TCA_PIE_BETA]); + + if (tb[TCA_PIE_ECN]) + q->params.ecn = nla_get_u32(tb[TCA_PIE_ECN]); + + if (tb[TCA_PIE_BYTEMODE]) + q->params.bytemode = nla_get_u32(tb[TCA_PIE_BYTEMODE]); + + /* Drop excess packets if new limit is lower */ + qlen = sch->q.qlen; + while (sch->q.qlen > sch->limit) { + struct sk_buff *skb = __qdisc_dequeue_head(&sch->q); + + dropped += qdisc_pkt_len(skb); + qdisc_qstats_backlog_dec(sch, skb); + rtnl_qdisc_drop(skb, sch); + } + qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped); + + sch_tree_unlock(sch); + return 0; +} + +static void pie_process_dequeue(struct Qdisc *sch, struct sk_buff *skb) +{ + + struct pie_sched_data *q = qdisc_priv(sch); + int qlen = sch->qstats.backlog; /* current queue size in bytes */ + + /* If current queue is about 10 packets or more and dq_count is unset + * we have enough packets to calculate the drain rate. Save + * current time as dq_tstamp and start measurement cycle. + */ + if (qlen >= QUEUE_THRESHOLD && q->vars.dq_count == DQCOUNT_INVALID) { + q->vars.dq_tstamp = psched_get_time(); + q->vars.dq_count = 0; + } + + /* Calculate the average drain rate from this value. If queue length + * has receded to a small value viz., <= QUEUE_THRESHOLD bytes,reset + * the dq_count to -1 as we don't have enough packets to calculate the + * drain rate anymore The following if block is entered only when we + * have a substantial queue built up (QUEUE_THRESHOLD bytes or more) + * and we calculate the drain rate for the threshold here. dq_count is + * in bytes, time difference in psched_time, hence rate is in + * bytes/psched_time. + */ + if (q->vars.dq_count != DQCOUNT_INVALID) { + q->vars.dq_count += skb->len; + + if (q->vars.dq_count >= QUEUE_THRESHOLD) { + psched_time_t now = psched_get_time(); + u32 dtime = now - q->vars.dq_tstamp; + u32 count = q->vars.dq_count << PIE_SCALE; + + if (dtime == 0) + return; + + count = count / dtime; + + if (q->vars.avg_dq_rate == 0) + q->vars.avg_dq_rate = count; + else + q->vars.avg_dq_rate = + (q->vars.avg_dq_rate - + (q->vars.avg_dq_rate >> 3)) + (count >> 3); + + /* If the queue has receded below the threshold, we hold + * on to the last drain rate calculated, else we reset + * dq_count to 0 to re-enter the if block when the next + * packet is dequeued + */ + if (qlen < QUEUE_THRESHOLD) + q->vars.dq_count = DQCOUNT_INVALID; + else { + q->vars.dq_count = 0; + q->vars.dq_tstamp = psched_get_time(); + } + + if (q->vars.burst_time > 0) { + if (q->vars.burst_time > dtime) + q->vars.burst_time -= dtime; + else + q->vars.burst_time = 0; + } + } + } +} + +static void calculate_probability(struct Qdisc *sch) +{ + struct pie_sched_data *q = qdisc_priv(sch); + u32 qlen = sch->qstats.backlog; /* queue size in bytes */ + psched_time_t qdelay = 0; /* in pschedtime */ + psched_time_t qdelay_old = q->vars.qdelay; /* in pschedtime */ + s32 delta = 0; /* determines the change in probability */ + u32 oldprob; + u32 alpha, beta; + bool update_prob = true; + + q->vars.qdelay_old = q->vars.qdelay; + + if (q->vars.avg_dq_rate > 0) + qdelay = (qlen << PIE_SCALE) / q->vars.avg_dq_rate; + else + qdelay = 0; + + /* If qdelay is zero and qlen is not, it means qlen is very small, less + * than dequeue_rate, so we do not update probabilty in this round + */ + if (qdelay == 0 && qlen != 0) + update_prob = false; + + /* In the algorithm, alpha and beta are between 0 and 2 with typical + * value for alpha as 0.125. In this implementation, we use values 0-32 + * passed from user space to represent this. Also, alpha and beta have + * unit of HZ and need to be scaled before they can used to update + * probability. alpha/beta are updated locally below by 1) scaling them + * appropriately 2) scaling down by 16 to come to 0-2 range. + * Please see paper for details. + * + * We scale alpha and beta differently depending on whether we are in + * light, medium or high dropping mode. + */ + if (q->vars.prob < MAX_PROB / 100) { + alpha = + (q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 7; + beta = + (q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 7; + } else if (q->vars.prob < MAX_PROB / 10) { + alpha = + (q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 5; + beta = + (q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 5; + } else { + alpha = + (q->params.alpha * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4; + beta = + (q->params.beta * (MAX_PROB / PSCHED_TICKS_PER_SEC)) >> 4; + } + + /* alpha and beta should be between 0 and 32, in multiples of 1/16 */ + delta += alpha * ((qdelay - q->params.target)); + delta += beta * ((qdelay - qdelay_old)); + + oldprob = q->vars.prob; + + /* to ensure we increase probability in steps of no more than 2% */ + if (delta > (s32) (MAX_PROB / (100 / 2)) && + q->vars.prob >= MAX_PROB / 10) + delta = (MAX_PROB / 100) * 2; + + /* Non-linear drop: + * Tune drop probability to increase quickly for high delays(>= 250ms) + * 250ms is derived through experiments and provides error protection + */ + + if (qdelay > (PSCHED_NS2TICKS(250 * NSEC_PER_MSEC))) + delta += MAX_PROB / (100 / 2); + + q->vars.prob += delta; + + if (delta > 0) { + /* prevent overflow */ + if (q->vars.prob < oldprob) { + q->vars.prob = MAX_PROB; + /* Prevent normalization error. If probability is at + * maximum value already, we normalize it here, and + * skip the check to do a non-linear drop in the next + * section. + */ + update_prob = false; + } + } else { + /* prevent underflow */ + if (q->vars.prob > oldprob) + q->vars.prob = 0; + } + + /* Non-linear drop in probability: Reduce drop probability quickly if + * delay is 0 for 2 consecutive Tupdate periods. + */ + + if ((qdelay == 0) && (qdelay_old == 0) && update_prob) + q->vars.prob = (q->vars.prob * 98) / 100; + + q->vars.qdelay = qdelay; + q->vars.qlen_old = qlen; + + /* We restart the measurement cycle if the following conditions are met + * 1. If the delay has been low for 2 consecutive Tupdate periods + * 2. Calculated drop probability is zero + * 3. We have atleast one estimate for the avg_dq_rate ie., + * is a non-zero value + */ + if ((q->vars.qdelay < q->params.target / 2) && + (q->vars.qdelay_old < q->params.target / 2) && + (q->vars.prob == 0) && + (q->vars.avg_dq_rate > 0)) + pie_vars_init(&q->vars); +} + +static void pie_timer(struct timer_list *t) +{ + struct pie_sched_data *q = from_timer(q, t, adapt_timer); + struct Qdisc *sch = q->sch; + spinlock_t *root_lock = qdisc_lock(qdisc_root_sleeping(sch)); + + spin_lock(root_lock); + calculate_probability(sch); + + /* reset the timer to fire after 'tupdate'. tupdate is in jiffies. */ + if (q->params.tupdate) + mod_timer(&q->adapt_timer, jiffies + q->params.tupdate); + spin_unlock(root_lock); + +} + +static int pie_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct pie_sched_data *q = qdisc_priv(sch); + + pie_params_init(&q->params); + pie_vars_init(&q->vars); + sch->limit = q->params.limit; + + q->sch = sch; + timer_setup(&q->adapt_timer, pie_timer, 0); + + if (opt) { + int err = pie_change(sch, opt, extack); + + if (err) + return err; + } + + mod_timer(&q->adapt_timer, jiffies + HZ / 2); + return 0; +} + +static int pie_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct pie_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + + /* convert target from pschedtime to us */ + if (nla_put_u32(skb, TCA_PIE_TARGET, + ((u32) PSCHED_TICKS2NS(q->params.target)) / + NSEC_PER_USEC) || + nla_put_u32(skb, TCA_PIE_LIMIT, sch->limit) || + nla_put_u32(skb, TCA_PIE_TUPDATE, jiffies_to_usecs(q->params.tupdate)) || + nla_put_u32(skb, TCA_PIE_ALPHA, q->params.alpha) || + nla_put_u32(skb, TCA_PIE_BETA, q->params.beta) || + nla_put_u32(skb, TCA_PIE_ECN, q->params.ecn) || + nla_put_u32(skb, TCA_PIE_BYTEMODE, q->params.bytemode)) + goto nla_put_failure; + + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -1; + +} + +static int pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct pie_sched_data *q = qdisc_priv(sch); + struct tc_pie_xstats st = { + .prob = q->vars.prob, + .delay = ((u32) PSCHED_TICKS2NS(q->vars.qdelay)) / + NSEC_PER_USEC, + /* unscale and return dq_rate in bytes per sec */ + .avg_dq_rate = q->vars.avg_dq_rate * + (PSCHED_TICKS_PER_SEC) >> PIE_SCALE, + .packets_in = q->stats.packets_in, + .overlimit = q->stats.overlimit, + .maxq = q->stats.maxq, + .dropped = q->stats.dropped, + .ecn_mark = q->stats.ecn_mark, + }; + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static struct sk_buff *pie_qdisc_dequeue(struct Qdisc *sch) +{ + struct sk_buff *skb; + skb = qdisc_dequeue_head(sch); + + if (!skb) + return NULL; + + pie_process_dequeue(sch, skb); + return skb; +} + +static void pie_reset(struct Qdisc *sch) +{ + struct pie_sched_data *q = qdisc_priv(sch); + qdisc_reset_queue(sch); + pie_vars_init(&q->vars); +} + +static void pie_destroy(struct Qdisc *sch) +{ + struct pie_sched_data *q = qdisc_priv(sch); + q->params.tupdate = 0; + del_timer_sync(&q->adapt_timer); +} + +static struct Qdisc_ops pie_qdisc_ops __read_mostly = { + .id = "pie", + .priv_size = sizeof(struct pie_sched_data), + .enqueue = pie_qdisc_enqueue, + .dequeue = pie_qdisc_dequeue, + .peek = qdisc_peek_dequeued, + .init = pie_init, + .destroy = pie_destroy, + .reset = pie_reset, + .change = pie_change, + .dump = pie_dump, + .dump_stats = pie_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init pie_module_init(void) +{ + return register_qdisc(&pie_qdisc_ops); +} + +static void __exit pie_module_exit(void) +{ + unregister_qdisc(&pie_qdisc_ops); +} + +module_init(pie_module_init); +module_exit(pie_module_exit); + +MODULE_DESCRIPTION("Proportional Integral controller Enhanced (PIE) scheduler"); +MODULE_AUTHOR("Vijay Subramanian"); +MODULE_AUTHOR("Mythili Prabhu"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_plug.c b/net/sched/sch_plug.c new file mode 100644 index 000000000..5619d2eb1 --- /dev/null +++ b/net/sched/sch_plug.c @@ -0,0 +1,235 @@ +/* + * sch_plug.c Queue traffic until an explicit release command + * + * 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. + * + * There are two ways to use this qdisc: + * 1. A simple "instantaneous" plug/unplug operation, by issuing an alternating + * sequence of TCQ_PLUG_BUFFER & TCQ_PLUG_RELEASE_INDEFINITE commands. + * + * 2. For network output buffering (a.k.a output commit) functionality. + * Output commit property is commonly used by applications using checkpoint + * based fault-tolerance to ensure that the checkpoint from which a system + * is being restored is consistent w.r.t outside world. + * + * Consider for e.g. Remus - a Virtual Machine checkpointing system, + * wherein a VM is checkpointed, say every 50ms. The checkpoint is replicated + * asynchronously to the backup host, while the VM continues executing the + * next epoch speculatively. + * + * The following is a typical sequence of output buffer operations: + * 1.At epoch i, start_buffer(i) + * 2. At end of epoch i (i.e. after 50ms): + * 2.1 Stop VM and take checkpoint(i). + * 2.2 start_buffer(i+1) and Resume VM + * 3. While speculatively executing epoch(i+1), asynchronously replicate + * checkpoint(i) to backup host. + * 4. When checkpoint_ack(i) is received from backup, release_buffer(i) + * Thus, this Qdisc would receive the following sequence of commands: + * TCQ_PLUG_BUFFER (epoch i) + * .. TCQ_PLUG_BUFFER (epoch i+1) + * ....TCQ_PLUG_RELEASE_ONE (epoch i) + * ......TCQ_PLUG_BUFFER (epoch i+2) + * ........ + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> + +/* + * State of the queue, when used for network output buffering: + * + * plug(i+1) plug(i) head + * ------------------+--------------------+----------------> + * | | + * | | + * pkts_current_epoch| pkts_last_epoch |pkts_to_release + * ----------------->|<--------+--------->|+---------------> + * v v + * + */ + +struct plug_sched_data { + /* If true, the dequeue function releases all packets + * from head to end of the queue. The queue turns into + * a pass-through queue for newly arriving packets. + */ + bool unplug_indefinite; + + bool throttled; + + /* Queue Limit in bytes */ + u32 limit; + + /* Number of packets (output) from the current speculatively + * executing epoch. + */ + u32 pkts_current_epoch; + + /* Number of packets corresponding to the recently finished + * epoch. These will be released when we receive a + * TCQ_PLUG_RELEASE_ONE command. This command is typically + * issued after committing a checkpoint at the target. + */ + u32 pkts_last_epoch; + + /* + * Number of packets from the head of the queue, that can + * be released (committed checkpoint). + */ + u32 pkts_to_release; +}; + +static int plug_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct plug_sched_data *q = qdisc_priv(sch); + + if (likely(sch->qstats.backlog + skb->len <= q->limit)) { + if (!q->unplug_indefinite) + q->pkts_current_epoch++; + return qdisc_enqueue_tail(skb, sch); + } + + return qdisc_drop(skb, sch, to_free); +} + +static struct sk_buff *plug_dequeue(struct Qdisc *sch) +{ + struct plug_sched_data *q = qdisc_priv(sch); + + if (q->throttled) + return NULL; + + if (!q->unplug_indefinite) { + if (!q->pkts_to_release) { + /* No more packets to dequeue. Block the queue + * and wait for the next release command. + */ + q->throttled = true; + return NULL; + } + q->pkts_to_release--; + } + + return qdisc_dequeue_head(sch); +} + +static int plug_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct plug_sched_data *q = qdisc_priv(sch); + + q->pkts_current_epoch = 0; + q->pkts_last_epoch = 0; + q->pkts_to_release = 0; + q->unplug_indefinite = false; + + if (opt == NULL) { + q->limit = qdisc_dev(sch)->tx_queue_len + * psched_mtu(qdisc_dev(sch)); + } else { + struct tc_plug_qopt *ctl = nla_data(opt); + + if (nla_len(opt) < sizeof(*ctl)) + return -EINVAL; + + q->limit = ctl->limit; + } + + q->throttled = true; + return 0; +} + +/* Receives 4 types of messages: + * TCQ_PLUG_BUFFER: Inset a plug into the queue and + * buffer any incoming packets + * TCQ_PLUG_RELEASE_ONE: Dequeue packets from queue head + * to beginning of the next plug. + * TCQ_PLUG_RELEASE_INDEFINITE: Dequeue all packets from queue. + * Stop buffering packets until the next TCQ_PLUG_BUFFER + * command is received (just act as a pass-thru queue). + * TCQ_PLUG_LIMIT: Increase/decrease queue size + */ +static int plug_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct plug_sched_data *q = qdisc_priv(sch); + struct tc_plug_qopt *msg; + + if (opt == NULL) + return -EINVAL; + + msg = nla_data(opt); + if (nla_len(opt) < sizeof(*msg)) + return -EINVAL; + + switch (msg->action) { + case TCQ_PLUG_BUFFER: + /* Save size of the current buffer */ + q->pkts_last_epoch = q->pkts_current_epoch; + q->pkts_current_epoch = 0; + if (q->unplug_indefinite) + q->throttled = true; + q->unplug_indefinite = false; + break; + case TCQ_PLUG_RELEASE_ONE: + /* Add packets from the last complete buffer to the + * packets to be released set. + */ + q->pkts_to_release += q->pkts_last_epoch; + q->pkts_last_epoch = 0; + q->throttled = false; + netif_schedule_queue(sch->dev_queue); + break; + case TCQ_PLUG_RELEASE_INDEFINITE: + q->unplug_indefinite = true; + q->pkts_to_release = 0; + q->pkts_last_epoch = 0; + q->pkts_current_epoch = 0; + q->throttled = false; + netif_schedule_queue(sch->dev_queue); + break; + case TCQ_PLUG_LIMIT: + /* Limit is supplied in bytes */ + q->limit = msg->limit; + break; + default: + return -EINVAL; + } + + return 0; +} + +static struct Qdisc_ops plug_qdisc_ops __read_mostly = { + .id = "plug", + .priv_size = sizeof(struct plug_sched_data), + .enqueue = plug_enqueue, + .dequeue = plug_dequeue, + .peek = qdisc_peek_head, + .init = plug_init, + .change = plug_change, + .reset = qdisc_reset_queue, + .owner = THIS_MODULE, +}; + +static int __init plug_module_init(void) +{ + return register_qdisc(&plug_qdisc_ops); +} + +static void __exit plug_module_exit(void) +{ + unregister_qdisc(&plug_qdisc_ops); +} +module_init(plug_module_init) +module_exit(plug_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c new file mode 100644 index 000000000..0e6f34bd9 --- /dev/null +++ b/net/sched/sch_prio.c @@ -0,0 +1,484 @@ +/* + * net/sched/sch_prio.c Simple 3-band priority "scheduler". + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * Fixes: 19990609: J Hadi Salim <hadi@nortelnetworks.com>: + * Init -- EINVAL when opt undefined + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + +struct prio_sched_data { + int bands; + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + u8 prio2band[TC_PRIO_MAX+1]; + struct Qdisc *queues[TCQ_PRIO_BANDS]; +}; + + +static struct Qdisc * +prio_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) +{ + struct prio_sched_data *q = qdisc_priv(sch); + u32 band = skb->priority; + struct tcf_result res; + struct tcf_proto *fl; + int err; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + if (TC_H_MAJ(skb->priority) != sch->handle) { + fl = rcu_dereference_bh(q->filter_list); + err = tcf_classify(skb, fl, &res, false); +#ifdef CONFIG_NET_CLS_ACT + switch (err) { + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + } +#endif + if (!fl || err < 0) { + if (TC_H_MAJ(band)) + band = 0; + return q->queues[q->prio2band[band & TC_PRIO_MAX]]; + } + band = res.classid; + } + band = TC_H_MIN(band) - 1; + if (band >= q->bands) + return q->queues[q->prio2band[0]]; + + return q->queues[band]; +} + +static int +prio_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) +{ + struct Qdisc *qdisc; + int ret; + + qdisc = prio_classify(skb, sch, &ret); +#ifdef CONFIG_NET_CLS_ACT + if (qdisc == NULL) { + + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; + } +#endif + + ret = qdisc_enqueue(skb, qdisc, to_free); + if (ret == NET_XMIT_SUCCESS) { + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + return NET_XMIT_SUCCESS; + } + if (net_xmit_drop_count(ret)) + qdisc_qstats_drop(sch); + return ret; +} + +static struct sk_buff *prio_peek(struct Qdisc *sch) +{ + struct prio_sched_data *q = qdisc_priv(sch); + int prio; + + for (prio = 0; prio < q->bands; prio++) { + struct Qdisc *qdisc = q->queues[prio]; + struct sk_buff *skb = qdisc->ops->peek(qdisc); + if (skb) + return skb; + } + return NULL; +} + +static struct sk_buff *prio_dequeue(struct Qdisc *sch) +{ + struct prio_sched_data *q = qdisc_priv(sch); + int prio; + + for (prio = 0; prio < q->bands; prio++) { + struct Qdisc *qdisc = q->queues[prio]; + struct sk_buff *skb = qdisc_dequeue_peeked(qdisc); + if (skb) { + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + return skb; + } + } + return NULL; + +} + +static void +prio_reset(struct Qdisc *sch) +{ + int prio; + struct prio_sched_data *q = qdisc_priv(sch); + + for (prio = 0; prio < q->bands; prio++) + qdisc_reset(q->queues[prio]); + sch->qstats.backlog = 0; + sch->q.qlen = 0; +} + +static int prio_offload(struct Qdisc *sch, struct tc_prio_qopt *qopt) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_prio_qopt_offload opt = { + .handle = sch->handle, + .parent = sch->parent, + }; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return -EOPNOTSUPP; + + if (qopt) { + opt.command = TC_PRIO_REPLACE; + opt.replace_params.bands = qopt->bands; + memcpy(&opt.replace_params.priomap, qopt->priomap, + TC_PRIO_MAX + 1); + opt.replace_params.qstats = &sch->qstats; + } else { + opt.command = TC_PRIO_DESTROY; + } + + return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_PRIO, &opt); +} + +static void +prio_destroy(struct Qdisc *sch) +{ + int prio; + struct prio_sched_data *q = qdisc_priv(sch); + + tcf_block_put(q->block); + prio_offload(sch, NULL); + for (prio = 0; prio < q->bands; prio++) + qdisc_put(q->queues[prio]); +} + +static int prio_tune(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct prio_sched_data *q = qdisc_priv(sch); + struct Qdisc *queues[TCQ_PRIO_BANDS]; + int oldbands = q->bands, i; + struct tc_prio_qopt *qopt; + + if (nla_len(opt) < sizeof(*qopt)) + return -EINVAL; + qopt = nla_data(opt); + + if (qopt->bands > TCQ_PRIO_BANDS || qopt->bands < 2) + return -EINVAL; + + for (i = 0; i <= TC_PRIO_MAX; i++) { + if (qopt->priomap[i] >= qopt->bands) + return -EINVAL; + } + + /* Before commit, make sure we can allocate all new qdiscs */ + for (i = oldbands; i < qopt->bands; i++) { + queues[i] = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + TC_H_MAKE(sch->handle, i + 1), + extack); + if (!queues[i]) { + while (i > oldbands) + qdisc_put(queues[--i]); + return -ENOMEM; + } + } + + prio_offload(sch, qopt); + sch_tree_lock(sch); + q->bands = qopt->bands; + memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1); + + for (i = q->bands; i < oldbands; i++) { + struct Qdisc *child = q->queues[i]; + + qdisc_tree_reduce_backlog(child, child->q.qlen, + child->qstats.backlog); + qdisc_put(child); + } + + for (i = oldbands; i < q->bands; i++) { + q->queues[i] = queues[i]; + if (q->queues[i] != &noop_qdisc) + qdisc_hash_add(q->queues[i], true); + } + + sch_tree_unlock(sch); + return 0; +} + +static int prio_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct prio_sched_data *q = qdisc_priv(sch); + int err; + + if (!opt) + return -EINVAL; + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + return prio_tune(sch, opt, extack); +} + +static int prio_dump_offload(struct Qdisc *sch) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_prio_qopt_offload hw_stats = { + .command = TC_PRIO_STATS, + .handle = sch->handle, + .parent = sch->parent, + { + .stats = { + .bstats = &sch->bstats, + .qstats = &sch->qstats, + }, + }, + }; + int err; + + sch->flags &= ~TCQ_F_OFFLOADED; + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return 0; + + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_PRIO, + &hw_stats); + if (err == -EOPNOTSUPP) + return 0; + + if (!err) + sch->flags |= TCQ_F_OFFLOADED; + + return err; +} + +static int prio_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct prio_sched_data *q = qdisc_priv(sch); + unsigned char *b = skb_tail_pointer(skb); + struct tc_prio_qopt opt; + int err; + + opt.bands = q->bands; + memcpy(&opt.priomap, q->prio2band, TC_PRIO_MAX + 1); + + err = prio_dump_offload(sch); + if (err) + goto nla_put_failure; + + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static int prio_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct prio_sched_data *q = qdisc_priv(sch); + struct tc_prio_qopt_offload graft_offload; + struct net_device *dev = qdisc_dev(sch); + unsigned long band = arg - 1; + bool any_qdisc_is_offloaded; + int err; + + if (!new) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + TC_H_MAKE(sch->handle, arg), extack); + if (!new) + new = &noop_qdisc; + else + qdisc_hash_add(new, true); + } + + *old = qdisc_replace(sch, new, &q->queues[band]); + + if (!tc_can_offload(dev)) + return 0; + + graft_offload.handle = sch->handle; + graft_offload.parent = sch->parent; + graft_offload.graft_params.band = band; + graft_offload.graft_params.child_handle = new->handle; + graft_offload.command = TC_PRIO_GRAFT; + + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_PRIO, + &graft_offload); + + /* Don't report error if the graft is part of destroy operation. */ + if (err && new != &noop_qdisc) { + /* Don't report error if the parent, the old child and the new + * one are not offloaded. + */ + any_qdisc_is_offloaded = sch->flags & TCQ_F_OFFLOADED; + any_qdisc_is_offloaded |= new->flags & TCQ_F_OFFLOADED; + if (*old) + any_qdisc_is_offloaded |= (*old)->flags & + TCQ_F_OFFLOADED; + + if (any_qdisc_is_offloaded) + NL_SET_ERR_MSG(extack, "Offloading graft operation failed."); + } + + return 0; +} + +static struct Qdisc * +prio_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct prio_sched_data *q = qdisc_priv(sch); + unsigned long band = arg - 1; + + return q->queues[band]; +} + +static unsigned long prio_find(struct Qdisc *sch, u32 classid) +{ + struct prio_sched_data *q = qdisc_priv(sch); + unsigned long band = TC_H_MIN(classid); + + if (band - 1 >= q->bands) + return 0; + return band; +} + +static unsigned long prio_bind(struct Qdisc *sch, unsigned long parent, u32 classid) +{ + return prio_find(sch, classid); +} + + +static void prio_unbind(struct Qdisc *q, unsigned long cl) +{ +} + +static int prio_dump_class(struct Qdisc *sch, unsigned long cl, struct sk_buff *skb, + struct tcmsg *tcm) +{ + struct prio_sched_data *q = qdisc_priv(sch); + + tcm->tcm_handle |= TC_H_MIN(cl); + tcm->tcm_info = q->queues[cl-1]->handle; + return 0; +} + +static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct prio_sched_data *q = qdisc_priv(sch); + struct Qdisc *cl_q; + + cl_q = q->queues[cl - 1]; + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, cl_q->cpu_bstats, &cl_q->bstats) < 0 || + gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0) + return -1; + + return 0; +} + +static void prio_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct prio_sched_data *q = qdisc_priv(sch); + int prio; + + if (arg->stop) + return; + + for (prio = 0; prio < q->bands; prio++) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, prio + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } +} + +static struct tcf_block *prio_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct prio_sched_data *q = qdisc_priv(sch); + + if (cl) + return NULL; + return q->block; +} + +static const struct Qdisc_class_ops prio_class_ops = { + .graft = prio_graft, + .leaf = prio_leaf, + .find = prio_find, + .walk = prio_walk, + .tcf_block = prio_tcf_block, + .bind_tcf = prio_bind, + .unbind_tcf = prio_unbind, + .dump = prio_dump_class, + .dump_stats = prio_dump_class_stats, +}; + +static struct Qdisc_ops prio_qdisc_ops __read_mostly = { + .next = NULL, + .cl_ops = &prio_class_ops, + .id = "prio", + .priv_size = sizeof(struct prio_sched_data), + .enqueue = prio_enqueue, + .dequeue = prio_dequeue, + .peek = prio_peek, + .init = prio_init, + .reset = prio_reset, + .destroy = prio_destroy, + .change = prio_tune, + .dump = prio_dump, + .owner = THIS_MODULE, +}; + +static int __init prio_module_init(void) +{ + return register_qdisc(&prio_qdisc_ops); +} + +static void __exit prio_module_exit(void) +{ + unregister_qdisc(&prio_qdisc_ops); +} + +module_init(prio_module_init) +module_exit(prio_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c new file mode 100644 index 000000000..20dc1851d --- /dev/null +++ b/net/sched/sch_qfq.c @@ -0,0 +1,1532 @@ +/* + * net/sched/sch_qfq.c Quick Fair Queueing Plus Scheduler. + * + * Copyright (c) 2009 Fabio Checconi, Luigi Rizzo, and Paolo Valente. + * Copyright (c) 2012 Paolo Valente. + * + * 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/init.h> +#include <linux/bitops.h> +#include <linux/errno.h> +#include <linux/netdevice.h> +#include <linux/pkt_sched.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> + + +/* Quick Fair Queueing Plus + ======================== + + Sources: + + [1] Paolo Valente, + "Reducing the Execution Time of Fair-Queueing Schedulers." + http://algo.ing.unimo.it/people/paolo/agg-sched/agg-sched.pdf + + Sources for QFQ: + + [2] Fabio Checconi, Luigi Rizzo, and Paolo Valente: "QFQ: Efficient + Packet Scheduling with Tight Bandwidth Distribution Guarantees." + + See also: + http://retis.sssup.it/~fabio/linux/qfq/ + */ + +/* + + QFQ+ divides classes into aggregates of at most MAX_AGG_CLASSES + classes. Each aggregate is timestamped with a virtual start time S + and a virtual finish time F, and scheduled according to its + timestamps. S and F are computed as a function of a system virtual + time function V. The classes within each aggregate are instead + scheduled with DRR. + + To speed up operations, QFQ+ divides also aggregates into a limited + number of groups. Which group a class belongs to depends on the + ratio between the maximum packet length for the class and the weight + of the class. Groups have their own S and F. In the end, QFQ+ + schedules groups, then aggregates within groups, then classes within + aggregates. See [1] and [2] for a full description. + + Virtual time computations. + + S, F and V are all computed in fixed point arithmetic with + FRAC_BITS decimal bits. + + QFQ_MAX_INDEX is the maximum index allowed for a group. We need + one bit per index. + QFQ_MAX_WSHIFT is the maximum power of two supported as a weight. + + The layout of the bits is as below: + + [ MTU_SHIFT ][ FRAC_BITS ] + [ MAX_INDEX ][ MIN_SLOT_SHIFT ] + ^.__grp->index = 0 + *.__grp->slot_shift + + where MIN_SLOT_SHIFT is derived by difference from the others. + + The max group index corresponds to Lmax/w_min, where + Lmax=1<<MTU_SHIFT, w_min = 1 . + From this, and knowing how many groups (MAX_INDEX) we want, + we can derive the shift corresponding to each group. + + Because we often need to compute + F = S + len/w_i and V = V + len/wsum + instead of storing w_i store the value + inv_w = (1<<FRAC_BITS)/w_i + so we can do F = S + len * inv_w * wsum. + We use W_TOT in the formulas so we can easily move between + static and adaptive weight sum. + + The per-scheduler-instance data contain all the data structures + for the scheduler: bitmaps and bucket lists. + + */ + +/* + * Maximum number of consecutive slots occupied by backlogged classes + * inside a group. + */ +#define QFQ_MAX_SLOTS 32 + +/* + * Shifts used for aggregate<->group mapping. We allow class weights that are + * in the range [1, 2^MAX_WSHIFT], and we try to map each aggregate i to the + * group with the smallest index that can support the L_i / r_i configured + * for the classes in the aggregate. + * + * grp->index is the index of the group; and grp->slot_shift + * is the shift for the corresponding (scaled) sigma_i. + */ +#define QFQ_MAX_INDEX 24 +#define QFQ_MAX_WSHIFT 10 + +#define QFQ_MAX_WEIGHT (1<<QFQ_MAX_WSHIFT) /* see qfq_slot_insert */ +#define QFQ_MAX_WSUM (64*QFQ_MAX_WEIGHT) + +#define FRAC_BITS 30 /* fixed point arithmetic */ +#define ONE_FP (1UL << FRAC_BITS) + +#define QFQ_MTU_SHIFT 16 /* to support TSO/GSO */ +#define QFQ_MIN_LMAX 512 /* see qfq_slot_insert */ + +#define QFQ_MAX_AGG_CLASSES 8 /* max num classes per aggregate allowed */ + +/* + * Possible group states. These values are used as indexes for the bitmaps + * array of struct qfq_queue. + */ +enum qfq_state { ER, IR, EB, IB, QFQ_MAX_STATE }; + +struct qfq_group; + +struct qfq_aggregate; + +struct qfq_class { + struct Qdisc_class_common common; + + unsigned int filter_cnt; + + struct gnet_stats_basic_packed bstats; + struct gnet_stats_queue qstats; + struct net_rate_estimator __rcu *rate_est; + struct Qdisc *qdisc; + struct list_head alist; /* Link for active-classes list. */ + struct qfq_aggregate *agg; /* Parent aggregate. */ + int deficit; /* DRR deficit counter. */ +}; + +struct qfq_aggregate { + struct hlist_node next; /* Link for the slot list. */ + u64 S, F; /* flow timestamps (exact) */ + + /* group we belong to. In principle we would need the index, + * which is log_2(lmax/weight), but we never reference it + * directly, only the group. + */ + struct qfq_group *grp; + + /* these are copied from the flowset. */ + u32 class_weight; /* Weight of each class in this aggregate. */ + /* Max pkt size for the classes in this aggregate, DRR quantum. */ + int lmax; + + u32 inv_w; /* ONE_FP/(sum of weights of classes in aggr.). */ + u32 budgetmax; /* Max budget for this aggregate. */ + u32 initial_budget, budget; /* Initial and current budget. */ + + int num_classes; /* Number of classes in this aggr. */ + struct list_head active; /* DRR queue of active classes. */ + + struct hlist_node nonfull_next; /* See nonfull_aggs in qfq_sched. */ +}; + +struct qfq_group { + u64 S, F; /* group timestamps (approx). */ + unsigned int slot_shift; /* Slot shift. */ + unsigned int index; /* Group index. */ + unsigned int front; /* Index of the front slot. */ + unsigned long full_slots; /* non-empty slots */ + + /* Array of RR lists of active aggregates. */ + struct hlist_head slots[QFQ_MAX_SLOTS]; +}; + +struct qfq_sched { + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + struct Qdisc_class_hash clhash; + + u64 oldV, V; /* Precise virtual times. */ + struct qfq_aggregate *in_serv_agg; /* Aggregate being served. */ + u32 wsum; /* weight sum */ + u32 iwsum; /* inverse weight sum */ + + unsigned long bitmaps[QFQ_MAX_STATE]; /* Group bitmaps. */ + struct qfq_group groups[QFQ_MAX_INDEX + 1]; /* The groups. */ + u32 min_slot_shift; /* Index of the group-0 bit in the bitmaps. */ + + u32 max_agg_classes; /* Max number of classes per aggr. */ + struct hlist_head nonfull_aggs; /* Aggs with room for more classes. */ +}; + +/* + * Possible reasons why the timestamps of an aggregate are updated + * enqueue: the aggregate switches from idle to active and must scheduled + * for service + * requeue: the aggregate finishes its budget, so it stops being served and + * must be rescheduled for service + */ +enum update_reason {enqueue, requeue}; + +static struct qfq_class *qfq_find_class(struct Qdisc *sch, u32 classid) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct Qdisc_class_common *clc; + + clc = qdisc_class_find(&q->clhash, classid); + if (clc == NULL) + return NULL; + return container_of(clc, struct qfq_class, common); +} + +static void qfq_purge_queue(struct qfq_class *cl) +{ + unsigned int len = cl->qdisc->q.qlen; + unsigned int backlog = cl->qdisc->qstats.backlog; + + qdisc_reset(cl->qdisc); + qdisc_tree_reduce_backlog(cl->qdisc, len, backlog); +} + +static const struct nla_policy qfq_policy[TCA_QFQ_MAX + 1] = { + [TCA_QFQ_WEIGHT] = { .type = NLA_U32 }, + [TCA_QFQ_LMAX] = { .type = NLA_U32 }, +}; + +/* + * Calculate a flow index, given its weight and maximum packet length. + * index = log_2(maxlen/weight) but we need to apply the scaling. + * This is used only once at flow creation. + */ +static int qfq_calc_index(u32 inv_w, unsigned int maxlen, u32 min_slot_shift) +{ + u64 slot_size = (u64)maxlen * inv_w; + unsigned long size_map; + int index = 0; + + size_map = slot_size >> min_slot_shift; + if (!size_map) + goto out; + + index = __fls(size_map) + 1; /* basically a log_2 */ + index -= !(slot_size - (1ULL << (index + min_slot_shift - 1))); + + if (index < 0) + index = 0; +out: + pr_debug("qfq calc_index: W = %lu, L = %u, I = %d\n", + (unsigned long) ONE_FP/inv_w, maxlen, index); + + return index; +} + +static void qfq_deactivate_agg(struct qfq_sched *, struct qfq_aggregate *); +static void qfq_activate_agg(struct qfq_sched *, struct qfq_aggregate *, + enum update_reason); + +static void qfq_init_agg(struct qfq_sched *q, struct qfq_aggregate *agg, + u32 lmax, u32 weight) +{ + INIT_LIST_HEAD(&agg->active); + hlist_add_head(&agg->nonfull_next, &q->nonfull_aggs); + + agg->lmax = lmax; + agg->class_weight = weight; +} + +static struct qfq_aggregate *qfq_find_agg(struct qfq_sched *q, + u32 lmax, u32 weight) +{ + struct qfq_aggregate *agg; + + hlist_for_each_entry(agg, &q->nonfull_aggs, nonfull_next) + if (agg->lmax == lmax && agg->class_weight == weight) + return agg; + + return NULL; +} + + +/* Update aggregate as a function of the new number of classes. */ +static void qfq_update_agg(struct qfq_sched *q, struct qfq_aggregate *agg, + int new_num_classes) +{ + u32 new_agg_weight; + + if (new_num_classes == q->max_agg_classes) + hlist_del_init(&agg->nonfull_next); + + if (agg->num_classes > new_num_classes && + new_num_classes == q->max_agg_classes - 1) /* agg no more full */ + hlist_add_head(&agg->nonfull_next, &q->nonfull_aggs); + + /* The next assignment may let + * agg->initial_budget > agg->budgetmax + * hold, we will take it into account in charge_actual_service(). + */ + agg->budgetmax = new_num_classes * agg->lmax; + new_agg_weight = agg->class_weight * new_num_classes; + agg->inv_w = ONE_FP/new_agg_weight; + + if (agg->grp == NULL) { + int i = qfq_calc_index(agg->inv_w, agg->budgetmax, + q->min_slot_shift); + agg->grp = &q->groups[i]; + } + + q->wsum += + (int) agg->class_weight * (new_num_classes - agg->num_classes); + q->iwsum = ONE_FP / q->wsum; + + agg->num_classes = new_num_classes; +} + +/* Add class to aggregate. */ +static void qfq_add_to_agg(struct qfq_sched *q, + struct qfq_aggregate *agg, + struct qfq_class *cl) +{ + cl->agg = agg; + + qfq_update_agg(q, agg, agg->num_classes+1); + if (cl->qdisc->q.qlen > 0) { /* adding an active class */ + list_add_tail(&cl->alist, &agg->active); + if (list_first_entry(&agg->active, struct qfq_class, alist) == + cl && q->in_serv_agg != agg) /* agg was inactive */ + qfq_activate_agg(q, agg, enqueue); /* schedule agg */ + } +} + +static struct qfq_aggregate *qfq_choose_next_agg(struct qfq_sched *); + +static void qfq_destroy_agg(struct qfq_sched *q, struct qfq_aggregate *agg) +{ + hlist_del_init(&agg->nonfull_next); + q->wsum -= agg->class_weight; + if (q->wsum != 0) + q->iwsum = ONE_FP / q->wsum; + + if (q->in_serv_agg == agg) + q->in_serv_agg = qfq_choose_next_agg(q); + kfree(agg); +} + +/* Deschedule class from within its parent aggregate. */ +static void qfq_deactivate_class(struct qfq_sched *q, struct qfq_class *cl) +{ + struct qfq_aggregate *agg = cl->agg; + + + list_del(&cl->alist); /* remove from RR queue of the aggregate */ + if (list_empty(&agg->active)) /* agg is now inactive */ + qfq_deactivate_agg(q, agg); +} + +/* Remove class from its parent aggregate. */ +static void qfq_rm_from_agg(struct qfq_sched *q, struct qfq_class *cl) +{ + struct qfq_aggregate *agg = cl->agg; + + cl->agg = NULL; + if (agg->num_classes == 1) { /* agg being emptied, destroy it */ + qfq_destroy_agg(q, agg); + return; + } + qfq_update_agg(q, agg, agg->num_classes-1); +} + +/* Deschedule class and remove it from its parent aggregate. */ +static void qfq_deact_rm_from_agg(struct qfq_sched *q, struct qfq_class *cl) +{ + if (cl->qdisc->q.qlen > 0) /* class is active */ + qfq_deactivate_class(q, cl); + + qfq_rm_from_agg(q, cl); +} + +/* Move class to a new aggregate, matching the new class weight and/or lmax */ +static int qfq_change_agg(struct Qdisc *sch, struct qfq_class *cl, u32 weight, + u32 lmax) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_aggregate *new_agg = qfq_find_agg(q, lmax, weight); + + if (new_agg == NULL) { /* create new aggregate */ + new_agg = kzalloc(sizeof(*new_agg), GFP_ATOMIC); + if (new_agg == NULL) + return -ENOBUFS; + qfq_init_agg(q, new_agg, lmax, weight); + } + qfq_deact_rm_from_agg(q, cl); + qfq_add_to_agg(q, new_agg, cl); + + return 0; +} + +static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, + struct nlattr **tca, unsigned long *arg, + struct netlink_ext_ack *extack) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl = (struct qfq_class *)*arg; + bool existing = false; + struct nlattr *tb[TCA_QFQ_MAX + 1]; + struct qfq_aggregate *new_agg = NULL; + u32 weight, lmax, inv_w; + int err; + int delta_w; + + if (tca[TCA_OPTIONS] == NULL) { + pr_notice("qfq: no options\n"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_QFQ_MAX, tca[TCA_OPTIONS], qfq_policy, + NULL); + if (err < 0) + return err; + + if (tb[TCA_QFQ_WEIGHT]) { + weight = nla_get_u32(tb[TCA_QFQ_WEIGHT]); + if (!weight || weight > (1UL << QFQ_MAX_WSHIFT)) { + pr_notice("qfq: invalid weight %u\n", weight); + return -EINVAL; + } + } else + weight = 1; + + if (tb[TCA_QFQ_LMAX]) { + lmax = nla_get_u32(tb[TCA_QFQ_LMAX]); + if (lmax < QFQ_MIN_LMAX || lmax > (1UL << QFQ_MTU_SHIFT)) { + pr_notice("qfq: invalid max length %u\n", lmax); + return -EINVAL; + } + } else + lmax = psched_mtu(qdisc_dev(sch)); + + inv_w = ONE_FP / weight; + weight = ONE_FP / inv_w; + + if (cl != NULL && + lmax == cl->agg->lmax && + weight == cl->agg->class_weight) + return 0; /* nothing to change */ + + delta_w = weight - (cl ? cl->agg->class_weight : 0); + + if (q->wsum + delta_w > QFQ_MAX_WSUM) { + pr_notice("qfq: total weight out of range (%d + %u)\n", + delta_w, q->wsum); + return -EINVAL; + } + + if (cl != NULL) { /* modify existing class */ + if (tca[TCA_RATE]) { + err = gen_replace_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) + return err; + } + existing = true; + goto set_change_agg; + } + + /* create and init new class */ + cl = kzalloc(sizeof(struct qfq_class), GFP_KERNEL); + if (cl == NULL) + return -ENOBUFS; + + cl->common.classid = classid; + cl->deficit = lmax; + + cl->qdisc = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + classid, NULL); + if (cl->qdisc == NULL) + cl->qdisc = &noop_qdisc; + + if (tca[TCA_RATE]) { + err = gen_new_estimator(&cl->bstats, NULL, + &cl->rate_est, + NULL, + qdisc_root_sleeping_running(sch), + tca[TCA_RATE]); + if (err) + goto destroy_class; + } + + if (cl->qdisc != &noop_qdisc) + qdisc_hash_add(cl->qdisc, true); + +set_change_agg: + sch_tree_lock(sch); + new_agg = qfq_find_agg(q, lmax, weight); + if (new_agg == NULL) { /* create new aggregate */ + sch_tree_unlock(sch); + new_agg = kzalloc(sizeof(*new_agg), GFP_KERNEL); + if (new_agg == NULL) { + err = -ENOBUFS; + gen_kill_estimator(&cl->rate_est); + goto destroy_class; + } + sch_tree_lock(sch); + qfq_init_agg(q, new_agg, lmax, weight); + } + if (existing) + qfq_deact_rm_from_agg(q, cl); + else + qdisc_class_hash_insert(&q->clhash, &cl->common); + qfq_add_to_agg(q, new_agg, cl); + sch_tree_unlock(sch); + qdisc_class_hash_grow(sch, &q->clhash); + + *arg = (unsigned long)cl; + return 0; + +destroy_class: + qdisc_put(cl->qdisc); + kfree(cl); + return err; +} + +static void qfq_destroy_class(struct Qdisc *sch, struct qfq_class *cl) +{ + struct qfq_sched *q = qdisc_priv(sch); + + qfq_rm_from_agg(q, cl); + gen_kill_estimator(&cl->rate_est); + qdisc_put(cl->qdisc); + kfree(cl); +} + +static int qfq_delete_class(struct Qdisc *sch, unsigned long arg) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl = (struct qfq_class *)arg; + + if (cl->filter_cnt > 0) + return -EBUSY; + + sch_tree_lock(sch); + + qfq_purge_queue(cl); + qdisc_class_hash_remove(&q->clhash, &cl->common); + + sch_tree_unlock(sch); + + qfq_destroy_class(sch, cl); + return 0; +} + +static unsigned long qfq_search_class(struct Qdisc *sch, u32 classid) +{ + return (unsigned long)qfq_find_class(sch, classid); +} + +static struct tcf_block *qfq_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct qfq_sched *q = qdisc_priv(sch); + + if (cl) + return NULL; + + return q->block; +} + +static unsigned long qfq_bind_tcf(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + struct qfq_class *cl = qfq_find_class(sch, classid); + + if (cl != NULL) + cl->filter_cnt++; + + return (unsigned long)cl; +} + +static void qfq_unbind_tcf(struct Qdisc *sch, unsigned long arg) +{ + struct qfq_class *cl = (struct qfq_class *)arg; + + cl->filter_cnt--; +} + +static int qfq_graft_class(struct Qdisc *sch, unsigned long arg, + struct Qdisc *new, struct Qdisc **old, + struct netlink_ext_ack *extack) +{ + struct qfq_class *cl = (struct qfq_class *)arg; + + if (new == NULL) { + new = qdisc_create_dflt(sch->dev_queue, &pfifo_qdisc_ops, + cl->common.classid, NULL); + if (new == NULL) + new = &noop_qdisc; + } + + *old = qdisc_replace(sch, new, &cl->qdisc); + return 0; +} + +static struct Qdisc *qfq_class_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct qfq_class *cl = (struct qfq_class *)arg; + + return cl->qdisc; +} + +static int qfq_dump_class(struct Qdisc *sch, unsigned long arg, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct qfq_class *cl = (struct qfq_class *)arg; + struct nlattr *nest; + + tcm->tcm_parent = TC_H_ROOT; + tcm->tcm_handle = cl->common.classid; + tcm->tcm_info = cl->qdisc->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + if (nla_put_u32(skb, TCA_QFQ_WEIGHT, cl->agg->class_weight) || + nla_put_u32(skb, TCA_QFQ_LMAX, cl->agg->lmax)) + goto nla_put_failure; + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + +static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg, + struct gnet_dump *d) +{ + struct qfq_class *cl = (struct qfq_class *)arg; + struct tc_qfq_stats xstats; + + memset(&xstats, 0, sizeof(xstats)); + + xstats.weight = cl->agg->class_weight; + xstats.lmax = cl->agg->lmax; + + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || + gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || + gnet_stats_copy_queue(d, NULL, + &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0) + return -1; + + return gnet_stats_copy_app(d, &xstats, sizeof(xstats)); +} + +static void qfq_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl; + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, (unsigned long)cl, arg) < 0) { + arg->stop = 1; + return; + } + arg->count++; + } + } +} + +static struct qfq_class *qfq_classify(struct sk_buff *skb, struct Qdisc *sch, + int *qerr) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl; + struct tcf_result res; + struct tcf_proto *fl; + int result; + + if (TC_H_MAJ(skb->priority ^ sch->handle) == 0) { + pr_debug("qfq_classify: found %d\n", skb->priority); + cl = qfq_find_class(sch, skb->priority); + if (cl != NULL) + return cl; + } + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + fl = rcu_dereference_bh(q->filter_list); + result = tcf_classify(skb, fl, &res, false); + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_QUEUED: + case TC_ACT_STOLEN: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return NULL; + } +#endif + cl = (struct qfq_class *)res.class; + if (cl == NULL) + cl = qfq_find_class(sch, res.classid); + return cl; + } + + return NULL; +} + +/* Generic comparison function, handling wraparound. */ +static inline int qfq_gt(u64 a, u64 b) +{ + return (s64)(a - b) > 0; +} + +/* Round a precise timestamp to its slotted value. */ +static inline u64 qfq_round_down(u64 ts, unsigned int shift) +{ + return ts & ~((1ULL << shift) - 1); +} + +/* return the pointer to the group with lowest index in the bitmap */ +static inline struct qfq_group *qfq_ffs(struct qfq_sched *q, + unsigned long bitmap) +{ + int index = __ffs(bitmap); + return &q->groups[index]; +} +/* Calculate a mask to mimic what would be ffs_from(). */ +static inline unsigned long mask_from(unsigned long bitmap, int from) +{ + return bitmap & ~((1UL << from) - 1); +} + +/* + * The state computation relies on ER=0, IR=1, EB=2, IB=3 + * First compute eligibility comparing grp->S, q->V, + * then check if someone is blocking us and possibly add EB + */ +static int qfq_calc_state(struct qfq_sched *q, const struct qfq_group *grp) +{ + /* if S > V we are not eligible */ + unsigned int state = qfq_gt(grp->S, q->V); + unsigned long mask = mask_from(q->bitmaps[ER], grp->index); + struct qfq_group *next; + + if (mask) { + next = qfq_ffs(q, mask); + if (qfq_gt(grp->F, next->F)) + state |= EB; + } + + return state; +} + + +/* + * In principle + * q->bitmaps[dst] |= q->bitmaps[src] & mask; + * q->bitmaps[src] &= ~mask; + * but we should make sure that src != dst + */ +static inline void qfq_move_groups(struct qfq_sched *q, unsigned long mask, + int src, int dst) +{ + q->bitmaps[dst] |= q->bitmaps[src] & mask; + q->bitmaps[src] &= ~mask; +} + +static void qfq_unblock_groups(struct qfq_sched *q, int index, u64 old_F) +{ + unsigned long mask = mask_from(q->bitmaps[ER], index + 1); + struct qfq_group *next; + + if (mask) { + next = qfq_ffs(q, mask); + if (!qfq_gt(next->F, old_F)) + return; + } + + mask = (1UL << index) - 1; + qfq_move_groups(q, mask, EB, ER); + qfq_move_groups(q, mask, IB, IR); +} + +/* + * perhaps + * + old_V ^= q->V; + old_V >>= q->min_slot_shift; + if (old_V) { + ... + } + * + */ +static void qfq_make_eligible(struct qfq_sched *q) +{ + unsigned long vslot = q->V >> q->min_slot_shift; + unsigned long old_vslot = q->oldV >> q->min_slot_shift; + + if (vslot != old_vslot) { + unsigned long mask; + int last_flip_pos = fls(vslot ^ old_vslot); + + if (last_flip_pos > 31) /* higher than the number of groups */ + mask = ~0UL; /* make all groups eligible */ + else + mask = (1UL << last_flip_pos) - 1; + + qfq_move_groups(q, mask, IR, ER); + qfq_move_groups(q, mask, IB, EB); + } +} + +/* + * The index of the slot in which the input aggregate agg is to be + * inserted must not be higher than QFQ_MAX_SLOTS-2. There is a '-2' + * and not a '-1' because the start time of the group may be moved + * backward by one slot after the aggregate has been inserted, and + * this would cause non-empty slots to be right-shifted by one + * position. + * + * QFQ+ fully satisfies this bound to the slot index if the parameters + * of the classes are not changed dynamically, and if QFQ+ never + * happens to postpone the service of agg unjustly, i.e., it never + * happens that the aggregate becomes backlogged and eligible, or just + * eligible, while an aggregate with a higher approximated finish time + * is being served. In particular, in this case QFQ+ guarantees that + * the timestamps of agg are low enough that the slot index is never + * higher than 2. Unfortunately, QFQ+ cannot provide the same + * guarantee if it happens to unjustly postpone the service of agg, or + * if the parameters of some class are changed. + * + * As for the first event, i.e., an out-of-order service, the + * upper bound to the slot index guaranteed by QFQ+ grows to + * 2 + + * QFQ_MAX_AGG_CLASSES * ((1<<QFQ_MTU_SHIFT)/QFQ_MIN_LMAX) * + * (current_max_weight/current_wsum) <= 2 + 8 * 128 * 1. + * + * The following function deals with this problem by backward-shifting + * the timestamps of agg, if needed, so as to guarantee that the slot + * index is never higher than QFQ_MAX_SLOTS-2. This backward-shift may + * cause the service of other aggregates to be postponed, yet the + * worst-case guarantees of these aggregates are not violated. In + * fact, in case of no out-of-order service, the timestamps of agg + * would have been even lower than they are after the backward shift, + * because QFQ+ would have guaranteed a maximum value equal to 2 for + * the slot index, and 2 < QFQ_MAX_SLOTS-2. Hence the aggregates whose + * service is postponed because of the backward-shift would have + * however waited for the service of agg before being served. + * + * The other event that may cause the slot index to be higher than 2 + * for agg is a recent change of the parameters of some class. If the + * weight of a class is increased or the lmax (max_pkt_size) of the + * class is decreased, then a new aggregate with smaller slot size + * than the original parent aggregate of the class may happen to be + * activated. The activation of this aggregate should be properly + * delayed to when the service of the class has finished in the ideal + * system tracked by QFQ+. If the activation of the aggregate is not + * delayed to this reference time instant, then this aggregate may be + * unjustly served before other aggregates waiting for service. This + * may cause the above bound to the slot index to be violated for some + * of these unlucky aggregates. + * + * Instead of delaying the activation of the new aggregate, which is + * quite complex, the above-discussed capping of the slot index is + * used to handle also the consequences of a change of the parameters + * of a class. + */ +static void qfq_slot_insert(struct qfq_group *grp, struct qfq_aggregate *agg, + u64 roundedS) +{ + u64 slot = (roundedS - grp->S) >> grp->slot_shift; + unsigned int i; /* slot index in the bucket list */ + + if (unlikely(slot > QFQ_MAX_SLOTS - 2)) { + u64 deltaS = roundedS - grp->S - + ((u64)(QFQ_MAX_SLOTS - 2)<<grp->slot_shift); + agg->S -= deltaS; + agg->F -= deltaS; + slot = QFQ_MAX_SLOTS - 2; + } + + i = (grp->front + slot) % QFQ_MAX_SLOTS; + + hlist_add_head(&agg->next, &grp->slots[i]); + __set_bit(slot, &grp->full_slots); +} + +/* Maybe introduce hlist_first_entry?? */ +static struct qfq_aggregate *qfq_slot_head(struct qfq_group *grp) +{ + return hlist_entry(grp->slots[grp->front].first, + struct qfq_aggregate, next); +} + +/* + * remove the entry from the slot + */ +static void qfq_front_slot_remove(struct qfq_group *grp) +{ + struct qfq_aggregate *agg = qfq_slot_head(grp); + + BUG_ON(!agg); + hlist_del(&agg->next); + if (hlist_empty(&grp->slots[grp->front])) + __clear_bit(0, &grp->full_slots); +} + +/* + * Returns the first aggregate in the first non-empty bucket of the + * group. As a side effect, adjusts the bucket list so the first + * non-empty bucket is at position 0 in full_slots. + */ +static struct qfq_aggregate *qfq_slot_scan(struct qfq_group *grp) +{ + unsigned int i; + + pr_debug("qfq slot_scan: grp %u full %#lx\n", + grp->index, grp->full_slots); + + if (grp->full_slots == 0) + return NULL; + + i = __ffs(grp->full_slots); /* zero based */ + if (i > 0) { + grp->front = (grp->front + i) % QFQ_MAX_SLOTS; + grp->full_slots >>= i; + } + + return qfq_slot_head(grp); +} + +/* + * adjust the bucket list. When the start time of a group decreases, + * we move the index down (modulo QFQ_MAX_SLOTS) so we don't need to + * move the objects. The mask of occupied slots must be shifted + * because we use ffs() to find the first non-empty slot. + * This covers decreases in the group's start time, but what about + * increases of the start time ? + * Here too we should make sure that i is less than 32 + */ +static void qfq_slot_rotate(struct qfq_group *grp, u64 roundedS) +{ + unsigned int i = (grp->S - roundedS) >> grp->slot_shift; + + grp->full_slots <<= i; + grp->front = (grp->front - i) % QFQ_MAX_SLOTS; +} + +static void qfq_update_eligible(struct qfq_sched *q) +{ + struct qfq_group *grp; + unsigned long ineligible; + + ineligible = q->bitmaps[IR] | q->bitmaps[IB]; + if (ineligible) { + if (!q->bitmaps[ER]) { + grp = qfq_ffs(q, ineligible); + if (qfq_gt(grp->S, q->V)) + q->V = grp->S; + } + qfq_make_eligible(q); + } +} + +/* Dequeue head packet of the head class in the DRR queue of the aggregate. */ +static void agg_dequeue(struct qfq_aggregate *agg, + struct qfq_class *cl, unsigned int len) +{ + qdisc_dequeue_peeked(cl->qdisc); + + cl->deficit -= (int) len; + + if (cl->qdisc->q.qlen == 0) /* no more packets, remove from list */ + list_del(&cl->alist); + else if (cl->deficit < qdisc_pkt_len(cl->qdisc->ops->peek(cl->qdisc))) { + cl->deficit += agg->lmax; + list_move_tail(&cl->alist, &agg->active); + } +} + +static inline struct sk_buff *qfq_peek_skb(struct qfq_aggregate *agg, + struct qfq_class **cl, + unsigned int *len) +{ + struct sk_buff *skb; + + *cl = list_first_entry(&agg->active, struct qfq_class, alist); + skb = (*cl)->qdisc->ops->peek((*cl)->qdisc); + if (skb == NULL) + WARN_ONCE(1, "qfq_dequeue: non-workconserving leaf\n"); + else + *len = qdisc_pkt_len(skb); + + return skb; +} + +/* Update F according to the actual service received by the aggregate. */ +static inline void charge_actual_service(struct qfq_aggregate *agg) +{ + /* Compute the service received by the aggregate, taking into + * account that, after decreasing the number of classes in + * agg, it may happen that + * agg->initial_budget - agg->budget > agg->bugdetmax + */ + u32 service_received = min(agg->budgetmax, + agg->initial_budget - agg->budget); + + agg->F = agg->S + (u64)service_received * agg->inv_w; +} + +/* Assign a reasonable start time for a new aggregate in group i. + * Admissible values for \hat(F) are multiples of \sigma_i + * no greater than V+\sigma_i . Larger values mean that + * we had a wraparound so we consider the timestamp to be stale. + * + * If F is not stale and F >= V then we set S = F. + * Otherwise we should assign S = V, but this may violate + * the ordering in EB (see [2]). So, if we have groups in ER, + * set S to the F_j of the first group j which would be blocking us. + * We are guaranteed not to move S backward because + * otherwise our group i would still be blocked. + */ +static void qfq_update_start(struct qfq_sched *q, struct qfq_aggregate *agg) +{ + unsigned long mask; + u64 limit, roundedF; + int slot_shift = agg->grp->slot_shift; + + roundedF = qfq_round_down(agg->F, slot_shift); + limit = qfq_round_down(q->V, slot_shift) + (1ULL << slot_shift); + + if (!qfq_gt(agg->F, q->V) || qfq_gt(roundedF, limit)) { + /* timestamp was stale */ + mask = mask_from(q->bitmaps[ER], agg->grp->index); + if (mask) { + struct qfq_group *next = qfq_ffs(q, mask); + if (qfq_gt(roundedF, next->F)) { + if (qfq_gt(limit, next->F)) + agg->S = next->F; + else /* preserve timestamp correctness */ + agg->S = limit; + return; + } + } + agg->S = q->V; + } else /* timestamp is not stale */ + agg->S = agg->F; +} + +/* Update the timestamps of agg before scheduling/rescheduling it for + * service. In particular, assign to agg->F its maximum possible + * value, i.e., the virtual finish time with which the aggregate + * should be labeled if it used all its budget once in service. + */ +static inline void +qfq_update_agg_ts(struct qfq_sched *q, + struct qfq_aggregate *agg, enum update_reason reason) +{ + if (reason != requeue) + qfq_update_start(q, agg); + else /* just charge agg for the service received */ + agg->S = agg->F; + + agg->F = agg->S + (u64)agg->budgetmax * agg->inv_w; +} + +static void qfq_schedule_agg(struct qfq_sched *q, struct qfq_aggregate *agg); + +static struct sk_buff *qfq_dequeue(struct Qdisc *sch) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_aggregate *in_serv_agg = q->in_serv_agg; + struct qfq_class *cl; + struct sk_buff *skb = NULL; + /* next-packet len, 0 means no more active classes in in-service agg */ + unsigned int len = 0; + + if (in_serv_agg == NULL) + return NULL; + + if (!list_empty(&in_serv_agg->active)) + skb = qfq_peek_skb(in_serv_agg, &cl, &len); + + /* + * If there are no active classes in the in-service aggregate, + * or if the aggregate has not enough budget to serve its next + * class, then choose the next aggregate to serve. + */ + if (len == 0 || in_serv_agg->budget < len) { + charge_actual_service(in_serv_agg); + + /* recharge the budget of the aggregate */ + in_serv_agg->initial_budget = in_serv_agg->budget = + in_serv_agg->budgetmax; + + if (!list_empty(&in_serv_agg->active)) { + /* + * Still active: reschedule for + * service. Possible optimization: if no other + * aggregate is active, then there is no point + * in rescheduling this aggregate, and we can + * just keep it as the in-service one. This + * should be however a corner case, and to + * handle it, we would need to maintain an + * extra num_active_aggs field. + */ + qfq_update_agg_ts(q, in_serv_agg, requeue); + qfq_schedule_agg(q, in_serv_agg); + } else if (sch->q.qlen == 0) { /* no aggregate to serve */ + q->in_serv_agg = NULL; + return NULL; + } + + /* + * If we get here, there are other aggregates queued: + * choose the new aggregate to serve. + */ + in_serv_agg = q->in_serv_agg = qfq_choose_next_agg(q); + skb = qfq_peek_skb(in_serv_agg, &cl, &len); + } + if (!skb) + return NULL; + + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + qdisc_bstats_update(sch, skb); + + agg_dequeue(in_serv_agg, cl, len); + /* If lmax is lowered, through qfq_change_class, for a class + * owning pending packets with larger size than the new value + * of lmax, then the following condition may hold. + */ + if (unlikely(in_serv_agg->budget < len)) + in_serv_agg->budget = 0; + else + in_serv_agg->budget -= len; + + q->V += (u64)len * q->iwsum; + pr_debug("qfq dequeue: len %u F %lld now %lld\n", + len, (unsigned long long) in_serv_agg->F, + (unsigned long long) q->V); + + return skb; +} + +static struct qfq_aggregate *qfq_choose_next_agg(struct qfq_sched *q) +{ + struct qfq_group *grp; + struct qfq_aggregate *agg, *new_front_agg; + u64 old_F; + + qfq_update_eligible(q); + q->oldV = q->V; + + if (!q->bitmaps[ER]) + return NULL; + + grp = qfq_ffs(q, q->bitmaps[ER]); + old_F = grp->F; + + agg = qfq_slot_head(grp); + + /* agg starts to be served, remove it from schedule */ + qfq_front_slot_remove(grp); + + new_front_agg = qfq_slot_scan(grp); + + if (new_front_agg == NULL) /* group is now inactive, remove from ER */ + __clear_bit(grp->index, &q->bitmaps[ER]); + else { + u64 roundedS = qfq_round_down(new_front_agg->S, + grp->slot_shift); + unsigned int s; + + if (grp->S == roundedS) + return agg; + grp->S = roundedS; + grp->F = roundedS + (2ULL << grp->slot_shift); + __clear_bit(grp->index, &q->bitmaps[ER]); + s = qfq_calc_state(q, grp); + __set_bit(grp->index, &q->bitmaps[s]); + } + + qfq_unblock_groups(q, grp->index, old_F); + + return agg; +} + +static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl; + struct qfq_aggregate *agg; + int err = 0; + + cl = qfq_classify(skb, sch, &err); + if (cl == NULL) { + if (err & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return err; + } + pr_debug("qfq_enqueue: cl = %x\n", cl->common.classid); + + if (unlikely(cl->agg->lmax < qdisc_pkt_len(skb))) { + pr_debug("qfq: increasing maxpkt from %u to %u for class %u", + cl->agg->lmax, qdisc_pkt_len(skb), cl->common.classid); + err = qfq_change_agg(sch, cl, cl->agg->class_weight, + qdisc_pkt_len(skb)); + if (err) { + cl->qstats.drops++; + return qdisc_drop(skb, sch, to_free); + } + } + + err = qdisc_enqueue(skb, cl->qdisc, to_free); + if (unlikely(err != NET_XMIT_SUCCESS)) { + pr_debug("qfq_enqueue: enqueue failed %d\n", err); + if (net_xmit_drop_count(err)) { + cl->qstats.drops++; + qdisc_qstats_drop(sch); + } + return err; + } + + bstats_update(&cl->bstats, skb); + qdisc_qstats_backlog_inc(sch, skb); + ++sch->q.qlen; + + agg = cl->agg; + /* if the queue was not empty, then done here */ + if (cl->qdisc->q.qlen != 1) { + if (unlikely(skb == cl->qdisc->ops->peek(cl->qdisc)) && + list_first_entry(&agg->active, struct qfq_class, alist) + == cl && cl->deficit < qdisc_pkt_len(skb)) + list_move_tail(&cl->alist, &agg->active); + + return err; + } + + /* schedule class for service within the aggregate */ + cl->deficit = agg->lmax; + list_add_tail(&cl->alist, &agg->active); + + if (list_first_entry(&agg->active, struct qfq_class, alist) != cl || + q->in_serv_agg == agg) + return err; /* non-empty or in service, nothing else to do */ + + qfq_activate_agg(q, agg, enqueue); + + return err; +} + +/* + * Schedule aggregate according to its timestamps. + */ +static void qfq_schedule_agg(struct qfq_sched *q, struct qfq_aggregate *agg) +{ + struct qfq_group *grp = agg->grp; + u64 roundedS; + int s; + + roundedS = qfq_round_down(agg->S, grp->slot_shift); + + /* + * Insert agg in the correct bucket. + * If agg->S >= grp->S we don't need to adjust the + * bucket list and simply go to the insertion phase. + * Otherwise grp->S is decreasing, we must make room + * in the bucket list, and also recompute the group state. + * Finally, if there were no flows in this group and nobody + * was in ER make sure to adjust V. + */ + if (grp->full_slots) { + if (!qfq_gt(grp->S, agg->S)) + goto skip_update; + + /* create a slot for this agg->S */ + qfq_slot_rotate(grp, roundedS); + /* group was surely ineligible, remove */ + __clear_bit(grp->index, &q->bitmaps[IR]); + __clear_bit(grp->index, &q->bitmaps[IB]); + } else if (!q->bitmaps[ER] && qfq_gt(roundedS, q->V) && + q->in_serv_agg == NULL) + q->V = roundedS; + + grp->S = roundedS; + grp->F = roundedS + (2ULL << grp->slot_shift); + s = qfq_calc_state(q, grp); + __set_bit(grp->index, &q->bitmaps[s]); + + pr_debug("qfq enqueue: new state %d %#lx S %lld F %lld V %lld\n", + s, q->bitmaps[s], + (unsigned long long) agg->S, + (unsigned long long) agg->F, + (unsigned long long) q->V); + +skip_update: + qfq_slot_insert(grp, agg, roundedS); +} + + +/* Update agg ts and schedule agg for service */ +static void qfq_activate_agg(struct qfq_sched *q, struct qfq_aggregate *agg, + enum update_reason reason) +{ + agg->initial_budget = agg->budget = agg->budgetmax; /* recharge budg. */ + + qfq_update_agg_ts(q, agg, reason); + if (q->in_serv_agg == NULL) { /* no aggr. in service or scheduled */ + q->in_serv_agg = agg; /* start serving this aggregate */ + /* update V: to be in service, agg must be eligible */ + q->oldV = q->V = agg->S; + } else if (agg != q->in_serv_agg) + qfq_schedule_agg(q, agg); +} + +static void qfq_slot_remove(struct qfq_sched *q, struct qfq_group *grp, + struct qfq_aggregate *agg) +{ + unsigned int i, offset; + u64 roundedS; + + roundedS = qfq_round_down(agg->S, grp->slot_shift); + offset = (roundedS - grp->S) >> grp->slot_shift; + + i = (grp->front + offset) % QFQ_MAX_SLOTS; + + hlist_del(&agg->next); + if (hlist_empty(&grp->slots[i])) + __clear_bit(offset, &grp->full_slots); +} + +/* + * Called to forcibly deschedule an aggregate. If the aggregate is + * not in the front bucket, or if the latter has other aggregates in + * the front bucket, we can simply remove the aggregate with no other + * side effects. + * Otherwise we must propagate the event up. + */ +static void qfq_deactivate_agg(struct qfq_sched *q, struct qfq_aggregate *agg) +{ + struct qfq_group *grp = agg->grp; + unsigned long mask; + u64 roundedS; + int s; + + if (agg == q->in_serv_agg) { + charge_actual_service(agg); + q->in_serv_agg = qfq_choose_next_agg(q); + return; + } + + agg->F = agg->S; + qfq_slot_remove(q, grp, agg); + + if (!grp->full_slots) { + __clear_bit(grp->index, &q->bitmaps[IR]); + __clear_bit(grp->index, &q->bitmaps[EB]); + __clear_bit(grp->index, &q->bitmaps[IB]); + + if (test_bit(grp->index, &q->bitmaps[ER]) && + !(q->bitmaps[ER] & ~((1UL << grp->index) - 1))) { + mask = q->bitmaps[ER] & ((1UL << grp->index) - 1); + if (mask) + mask = ~((1UL << __fls(mask)) - 1); + else + mask = ~0UL; + qfq_move_groups(q, mask, EB, ER); + qfq_move_groups(q, mask, IB, IR); + } + __clear_bit(grp->index, &q->bitmaps[ER]); + } else if (hlist_empty(&grp->slots[grp->front])) { + agg = qfq_slot_scan(grp); + roundedS = qfq_round_down(agg->S, grp->slot_shift); + if (grp->S != roundedS) { + __clear_bit(grp->index, &q->bitmaps[ER]); + __clear_bit(grp->index, &q->bitmaps[IR]); + __clear_bit(grp->index, &q->bitmaps[EB]); + __clear_bit(grp->index, &q->bitmaps[IB]); + grp->S = roundedS; + grp->F = roundedS + (2ULL << grp->slot_shift); + s = qfq_calc_state(q, grp); + __set_bit(grp->index, &q->bitmaps[s]); + } + } +} + +static void qfq_qlen_notify(struct Qdisc *sch, unsigned long arg) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl = (struct qfq_class *)arg; + + qfq_deactivate_class(q, cl); +} + +static int qfq_init_qdisc(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_group *grp; + int i, j, err; + u32 max_cl_shift, maxbudg_shift, max_classes; + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + err = qdisc_class_hash_init(&q->clhash); + if (err < 0) + return err; + + max_classes = min_t(u64, (u64)qdisc_dev(sch)->tx_queue_len + 1, + QFQ_MAX_AGG_CLASSES); + /* max_cl_shift = floor(log_2(max_classes)) */ + max_cl_shift = __fls(max_classes); + q->max_agg_classes = 1<<max_cl_shift; + + /* maxbudg_shift = log2(max_len * max_classes_per_agg) */ + maxbudg_shift = QFQ_MTU_SHIFT + max_cl_shift; + q->min_slot_shift = FRAC_BITS + maxbudg_shift - QFQ_MAX_INDEX; + + for (i = 0; i <= QFQ_MAX_INDEX; i++) { + grp = &q->groups[i]; + grp->index = i; + grp->slot_shift = q->min_slot_shift + i; + for (j = 0; j < QFQ_MAX_SLOTS; j++) + INIT_HLIST_HEAD(&grp->slots[j]); + } + + INIT_HLIST_HEAD(&q->nonfull_aggs); + + return 0; +} + +static void qfq_reset_qdisc(struct Qdisc *sch) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl; + unsigned int i; + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { + if (cl->qdisc->q.qlen > 0) + qfq_deactivate_class(q, cl); + + qdisc_reset(cl->qdisc); + } + } + sch->qstats.backlog = 0; + sch->q.qlen = 0; +} + +static void qfq_destroy_qdisc(struct Qdisc *sch) +{ + struct qfq_sched *q = qdisc_priv(sch); + struct qfq_class *cl; + struct hlist_node *next; + unsigned int i; + + tcf_block_put(q->block); + + for (i = 0; i < q->clhash.hashsize; i++) { + hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], + common.hnode) { + qfq_destroy_class(sch, cl); + } + } + qdisc_class_hash_destroy(&q->clhash); +} + +static const struct Qdisc_class_ops qfq_class_ops = { + .change = qfq_change_class, + .delete = qfq_delete_class, + .find = qfq_search_class, + .tcf_block = qfq_tcf_block, + .bind_tcf = qfq_bind_tcf, + .unbind_tcf = qfq_unbind_tcf, + .graft = qfq_graft_class, + .leaf = qfq_class_leaf, + .qlen_notify = qfq_qlen_notify, + .dump = qfq_dump_class, + .dump_stats = qfq_dump_class_stats, + .walk = qfq_walk, +}; + +static struct Qdisc_ops qfq_qdisc_ops __read_mostly = { + .cl_ops = &qfq_class_ops, + .id = "qfq", + .priv_size = sizeof(struct qfq_sched), + .enqueue = qfq_enqueue, + .dequeue = qfq_dequeue, + .peek = qdisc_peek_dequeued, + .init = qfq_init_qdisc, + .reset = qfq_reset_qdisc, + .destroy = qfq_destroy_qdisc, + .owner = THIS_MODULE, +}; + +static int __init qfq_init(void) +{ + return register_qdisc(&qfq_qdisc_ops); +} + +static void __exit qfq_exit(void) +{ + unregister_qdisc(&qfq_qdisc_ops); +} + +module_init(qfq_init); +module_exit(qfq_exit); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c new file mode 100644 index 000000000..0424aa747 --- /dev/null +++ b/net/sched/sch_red.c @@ -0,0 +1,455 @@ +/* + * net/sched/sch_red.c Random Early Detection queue. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * + * Changes: + * J Hadi Salim 980914: computation fixes + * Alexey Makarenko <makar@phoenix.kharkov.ua> 990814: qave on idle link was calculated incorrectly. + * J Hadi Salim 980816: ECN support + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/inet_ecn.h> +#include <net/red.h> + + +/* Parameters, settable by user: + ----------------------------- + + limit - bytes (must be > qth_max + burst) + + Hard limit on queue length, should be chosen >qth_max + to allow packet bursts. This parameter does not + affect the algorithms behaviour and can be chosen + arbitrarily high (well, less than ram size) + Really, this limit will never be reached + if RED works correctly. + */ + +struct red_sched_data { + u32 limit; /* HARD maximal queue length */ + unsigned char flags; + struct timer_list adapt_timer; + struct Qdisc *sch; + struct red_parms parms; + struct red_vars vars; + struct red_stats stats; + struct Qdisc *qdisc; +}; + +static inline int red_use_ecn(struct red_sched_data *q) +{ + return q->flags & TC_RED_ECN; +} + +static inline int red_use_harddrop(struct red_sched_data *q) +{ + return q->flags & TC_RED_HARDDROP; +} + +static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct Qdisc *child = q->qdisc; + int ret; + + q->vars.qavg = red_calc_qavg(&q->parms, + &q->vars, + child->qstats.backlog); + + if (red_is_idling(&q->vars)) + red_end_of_idle_period(&q->vars); + + switch (red_action(&q->parms, &q->vars, q->vars.qavg)) { + case RED_DONT_MARK: + break; + + case RED_PROB_MARK: + qdisc_qstats_overlimit(sch); + if (!red_use_ecn(q) || !INET_ECN_set_ce(skb)) { + q->stats.prob_drop++; + goto congestion_drop; + } + + q->stats.prob_mark++; + break; + + case RED_HARD_MARK: + qdisc_qstats_overlimit(sch); + if (red_use_harddrop(q) || !red_use_ecn(q) || + !INET_ECN_set_ce(skb)) { + q->stats.forced_drop++; + goto congestion_drop; + } + + q->stats.forced_mark++; + break; + } + + ret = qdisc_enqueue(skb, child, to_free); + if (likely(ret == NET_XMIT_SUCCESS)) { + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + } else if (net_xmit_drop_count(ret)) { + q->stats.pdrop++; + qdisc_qstats_drop(sch); + } + return ret; + +congestion_drop: + qdisc_drop(skb, sch, to_free); + return NET_XMIT_CN; +} + +static struct sk_buff *red_dequeue(struct Qdisc *sch) +{ + struct sk_buff *skb; + struct red_sched_data *q = qdisc_priv(sch); + struct Qdisc *child = q->qdisc; + + skb = child->dequeue(child); + if (skb) { + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + } else { + if (!red_is_idling(&q->vars)) + red_start_of_idle_period(&q->vars); + } + return skb; +} + +static struct sk_buff *red_peek(struct Qdisc *sch) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct Qdisc *child = q->qdisc; + + return child->ops->peek(child); +} + +static void red_reset(struct Qdisc *sch) +{ + struct red_sched_data *q = qdisc_priv(sch); + + qdisc_reset(q->qdisc); + sch->qstats.backlog = 0; + sch->q.qlen = 0; + red_restart(&q->vars); +} + +static int red_offload(struct Qdisc *sch, bool enable) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + struct tc_red_qopt_offload opt = { + .handle = sch->handle, + .parent = sch->parent, + }; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return -EOPNOTSUPP; + + if (enable) { + opt.command = TC_RED_REPLACE; + opt.set.min = q->parms.qth_min >> q->parms.Wlog; + opt.set.max = q->parms.qth_max >> q->parms.Wlog; + opt.set.probability = q->parms.max_P; + opt.set.is_ecn = red_use_ecn(q); + opt.set.qstats = &sch->qstats; + } else { + opt.command = TC_RED_DESTROY; + } + + return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_RED, &opt); +} + +static void red_destroy(struct Qdisc *sch) +{ + struct red_sched_data *q = qdisc_priv(sch); + + del_timer_sync(&q->adapt_timer); + red_offload(sch, false); + qdisc_put(q->qdisc); +} + +static const struct nla_policy red_policy[TCA_RED_MAX + 1] = { + [TCA_RED_PARMS] = { .len = sizeof(struct tc_red_qopt) }, + [TCA_RED_STAB] = { .len = RED_STAB_SIZE }, + [TCA_RED_MAX_P] = { .type = NLA_U32 }, +}; + +static int red_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_RED_MAX + 1]; + struct tc_red_qopt *ctl; + struct Qdisc *child = NULL; + int err; + u32 max_P; + u8 *stab; + + if (opt == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_RED_MAX, opt, red_policy, NULL); + if (err < 0) + return err; + + if (tb[TCA_RED_PARMS] == NULL || + tb[TCA_RED_STAB] == NULL) + return -EINVAL; + + max_P = tb[TCA_RED_MAX_P] ? nla_get_u32(tb[TCA_RED_MAX_P]) : 0; + + ctl = nla_data(tb[TCA_RED_PARMS]); + stab = nla_data(tb[TCA_RED_STAB]); + if (!red_check_params(ctl->qth_min, ctl->qth_max, ctl->Wlog, + ctl->Scell_log, stab)) + return -EINVAL; + + if (ctl->limit > 0) { + child = fifo_create_dflt(sch, &bfifo_qdisc_ops, ctl->limit, + extack); + if (IS_ERR(child)) + return PTR_ERR(child); + + /* child is fifo, no need to check for noop_qdisc */ + qdisc_hash_add(child, true); + } + + sch_tree_lock(sch); + q->flags = ctl->flags; + q->limit = ctl->limit; + if (child) { + qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen, + q->qdisc->qstats.backlog); + qdisc_put(q->qdisc); + q->qdisc = child; + } + + red_set_parms(&q->parms, + ctl->qth_min, ctl->qth_max, ctl->Wlog, + ctl->Plog, ctl->Scell_log, + stab, + max_P); + red_set_vars(&q->vars); + + del_timer(&q->adapt_timer); + if (ctl->flags & TC_RED_ADAPTATIVE) + mod_timer(&q->adapt_timer, jiffies + HZ/2); + + if (!q->qdisc->q.qlen) + red_start_of_idle_period(&q->vars); + + sch_tree_unlock(sch); + red_offload(sch, true); + return 0; +} + +static inline void red_adaptative_timer(struct timer_list *t) +{ + struct red_sched_data *q = from_timer(q, t, adapt_timer); + struct Qdisc *sch = q->sch; + spinlock_t *root_lock = qdisc_lock(qdisc_root_sleeping(sch)); + + spin_lock(root_lock); + red_adaptative_algo(&q->parms, &q->vars); + mod_timer(&q->adapt_timer, jiffies + HZ/2); + spin_unlock(root_lock); +} + +static int red_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct red_sched_data *q = qdisc_priv(sch); + + q->qdisc = &noop_qdisc; + q->sch = sch; + timer_setup(&q->adapt_timer, red_adaptative_timer, 0); + return red_change(sch, opt, extack); +} + +static int red_dump_offload_stats(struct Qdisc *sch, struct tc_red_qopt *opt) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_red_qopt_offload hw_stats = { + .command = TC_RED_STATS, + .handle = sch->handle, + .parent = sch->parent, + { + .stats.bstats = &sch->bstats, + .stats.qstats = &sch->qstats, + }, + }; + int err; + + sch->flags &= ~TCQ_F_OFFLOADED; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return 0; + + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_RED, + &hw_stats); + if (err == -EOPNOTSUPP) + return 0; + + if (!err) + sch->flags |= TCQ_F_OFFLOADED; + + return err; +} + +static int red_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct nlattr *opts = NULL; + struct tc_red_qopt opt = { + .limit = q->limit, + .flags = q->flags, + .qth_min = q->parms.qth_min >> q->parms.Wlog, + .qth_max = q->parms.qth_max >> q->parms.Wlog, + .Wlog = q->parms.Wlog, + .Plog = q->parms.Plog, + .Scell_log = q->parms.Scell_log, + }; + int err; + + err = red_dump_offload_stats(sch, &opt); + if (err) + goto nla_put_failure; + + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + if (nla_put(skb, TCA_RED_PARMS, sizeof(opt), &opt) || + nla_put_u32(skb, TCA_RED_MAX_P, q->parms.max_P)) + goto nla_put_failure; + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -EMSGSIZE; +} + +static int red_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + struct tc_red_xstats st = {0}; + + if (sch->flags & TCQ_F_OFFLOADED) { + struct tc_red_qopt_offload hw_stats_request = { + .command = TC_RED_XSTATS, + .handle = sch->handle, + .parent = sch->parent, + { + .xstats = &q->stats, + }, + }; + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_RED, + &hw_stats_request); + } + st.early = q->stats.prob_drop + q->stats.forced_drop; + st.pdrop = q->stats.pdrop; + st.other = q->stats.other; + st.marked = q->stats.prob_mark + q->stats.forced_mark; + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static int red_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct red_sched_data *q = qdisc_priv(sch); + + tcm->tcm_handle |= TC_H_MIN(1); + tcm->tcm_info = q->qdisc->handle; + return 0; +} + +static int red_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct red_sched_data *q = qdisc_priv(sch); + + if (new == NULL) + new = &noop_qdisc; + + *old = qdisc_replace(sch, new, &q->qdisc); + return 0; +} + +static struct Qdisc *red_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct red_sched_data *q = qdisc_priv(sch); + return q->qdisc; +} + +static unsigned long red_find(struct Qdisc *sch, u32 classid) +{ + return 1; +} + +static void red_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + if (!walker->stop) { + if (walker->count >= walker->skip) + if (walker->fn(sch, 1, walker) < 0) { + walker->stop = 1; + return; + } + walker->count++; + } +} + +static const struct Qdisc_class_ops red_class_ops = { + .graft = red_graft, + .leaf = red_leaf, + .find = red_find, + .walk = red_walk, + .dump = red_dump_class, +}; + +static struct Qdisc_ops red_qdisc_ops __read_mostly = { + .id = "red", + .priv_size = sizeof(struct red_sched_data), + .cl_ops = &red_class_ops, + .enqueue = red_enqueue, + .dequeue = red_dequeue, + .peek = red_peek, + .init = red_init, + .reset = red_reset, + .destroy = red_destroy, + .change = red_change, + .dump = red_dump, + .dump_stats = red_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init red_module_init(void) +{ + return register_qdisc(&red_qdisc_ops); +} + +static void __exit red_module_exit(void) +{ + unregister_qdisc(&red_qdisc_ops); +} + +module_init(red_module_init) +module_exit(red_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c new file mode 100644 index 000000000..81d205acb --- /dev/null +++ b/net/sched/sch_sfb.c @@ -0,0 +1,734 @@ +/* + * net/sched/sch_sfb.c Stochastic Fair Blue + * + * Copyright (c) 2008-2011 Juliusz Chroboczek <jch@pps.jussieu.fr> + * Copyright (c) 2011 Eric Dumazet <eric.dumazet@gmail.com> + * + * 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. + * + * W. Feng, D. Kandlur, D. Saha, K. Shin. Blue: + * A New Class of Active Queue Management Algorithms. + * U. Michigan CSE-TR-387-99, April 1999. + * + * http://www.thefengs.com/wuchang/blue/CSE-TR-387-99.pdf + * + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/random.h> +#include <linux/siphash.h> +#include <net/ip.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/inet_ecn.h> + +/* + * SFB uses two B[l][n] : L x N arrays of bins (L levels, N bins per level) + * This implementation uses L = 8 and N = 16 + * This permits us to split one 32bit hash (provided per packet by rxhash or + * external classifier) into 8 subhashes of 4 bits. + */ +#define SFB_BUCKET_SHIFT 4 +#define SFB_NUMBUCKETS (1 << SFB_BUCKET_SHIFT) /* N bins per Level */ +#define SFB_BUCKET_MASK (SFB_NUMBUCKETS - 1) +#define SFB_LEVELS (32 / SFB_BUCKET_SHIFT) /* L */ + +/* SFB algo uses a virtual queue, named "bin" */ +struct sfb_bucket { + u16 qlen; /* length of virtual queue */ + u16 p_mark; /* marking probability */ +}; + +/* We use a double buffering right before hash change + * (Section 4.4 of SFB reference : moving hash functions) + */ +struct sfb_bins { + siphash_key_t perturbation; /* siphash key */ + struct sfb_bucket bins[SFB_LEVELS][SFB_NUMBUCKETS]; +}; + +struct sfb_sched_data { + struct Qdisc *qdisc; + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + unsigned long rehash_interval; + unsigned long warmup_time; /* double buffering warmup time in jiffies */ + u32 max; + u32 bin_size; /* maximum queue length per bin */ + u32 increment; /* d1 */ + u32 decrement; /* d2 */ + u32 limit; /* HARD maximal queue length */ + u32 penalty_rate; + u32 penalty_burst; + u32 tokens_avail; + unsigned long rehash_time; + unsigned long token_time; + + u8 slot; /* current active bins (0 or 1) */ + bool double_buffering; + struct sfb_bins bins[2]; + + struct { + u32 earlydrop; + u32 penaltydrop; + u32 bucketdrop; + u32 queuedrop; + u32 childdrop; /* drops in child qdisc */ + u32 marked; /* ECN mark */ + } stats; +}; + +/* + * Each queued skb might be hashed on one or two bins + * We store in skb_cb the two hash values. + * (A zero value means double buffering was not used) + */ +struct sfb_skb_cb { + u32 hashes[2]; +}; + +static inline struct sfb_skb_cb *sfb_skb_cb(const struct sk_buff *skb) +{ + qdisc_cb_private_validate(skb, sizeof(struct sfb_skb_cb)); + return (struct sfb_skb_cb *)qdisc_skb_cb(skb)->data; +} + +/* + * If using 'internal' SFB flow classifier, hash comes from skb rxhash + * If using external classifier, hash comes from the classid. + */ +static u32 sfb_hash(const struct sk_buff *skb, u32 slot) +{ + return sfb_skb_cb(skb)->hashes[slot]; +} + +/* Probabilities are coded as Q0.16 fixed-point values, + * with 0xFFFF representing 65535/65536 (almost 1.0) + * Addition and subtraction are saturating in [0, 65535] + */ +static u32 prob_plus(u32 p1, u32 p2) +{ + u32 res = p1 + p2; + + return min_t(u32, res, SFB_MAX_PROB); +} + +static u32 prob_minus(u32 p1, u32 p2) +{ + return p1 > p2 ? p1 - p2 : 0; +} + +static void increment_one_qlen(u32 sfbhash, u32 slot, struct sfb_sched_data *q) +{ + int i; + struct sfb_bucket *b = &q->bins[slot].bins[0][0]; + + for (i = 0; i < SFB_LEVELS; i++) { + u32 hash = sfbhash & SFB_BUCKET_MASK; + + sfbhash >>= SFB_BUCKET_SHIFT; + if (b[hash].qlen < 0xFFFF) + b[hash].qlen++; + b += SFB_NUMBUCKETS; /* next level */ + } +} + +static void increment_qlen(const struct sk_buff *skb, struct sfb_sched_data *q) +{ + u32 sfbhash; + + sfbhash = sfb_hash(skb, 0); + if (sfbhash) + increment_one_qlen(sfbhash, 0, q); + + sfbhash = sfb_hash(skb, 1); + if (sfbhash) + increment_one_qlen(sfbhash, 1, q); +} + +static void decrement_one_qlen(u32 sfbhash, u32 slot, + struct sfb_sched_data *q) +{ + int i; + struct sfb_bucket *b = &q->bins[slot].bins[0][0]; + + for (i = 0; i < SFB_LEVELS; i++) { + u32 hash = sfbhash & SFB_BUCKET_MASK; + + sfbhash >>= SFB_BUCKET_SHIFT; + if (b[hash].qlen > 0) + b[hash].qlen--; + b += SFB_NUMBUCKETS; /* next level */ + } +} + +static void decrement_qlen(const struct sk_buff *skb, struct sfb_sched_data *q) +{ + u32 sfbhash; + + sfbhash = sfb_hash(skb, 0); + if (sfbhash) + decrement_one_qlen(sfbhash, 0, q); + + sfbhash = sfb_hash(skb, 1); + if (sfbhash) + decrement_one_qlen(sfbhash, 1, q); +} + +static void decrement_prob(struct sfb_bucket *b, struct sfb_sched_data *q) +{ + b->p_mark = prob_minus(b->p_mark, q->decrement); +} + +static void increment_prob(struct sfb_bucket *b, struct sfb_sched_data *q) +{ + b->p_mark = prob_plus(b->p_mark, q->increment); +} + +static void sfb_zero_all_buckets(struct sfb_sched_data *q) +{ + memset(&q->bins, 0, sizeof(q->bins)); +} + +/* + * compute max qlen, max p_mark, and avg p_mark + */ +static u32 sfb_compute_qlen(u32 *prob_r, u32 *avgpm_r, const struct sfb_sched_data *q) +{ + int i; + u32 qlen = 0, prob = 0, totalpm = 0; + const struct sfb_bucket *b = &q->bins[q->slot].bins[0][0]; + + for (i = 0; i < SFB_LEVELS * SFB_NUMBUCKETS; i++) { + if (qlen < b->qlen) + qlen = b->qlen; + totalpm += b->p_mark; + if (prob < b->p_mark) + prob = b->p_mark; + b++; + } + *prob_r = prob; + *avgpm_r = totalpm / (SFB_LEVELS * SFB_NUMBUCKETS); + return qlen; +} + + +static void sfb_init_perturbation(u32 slot, struct sfb_sched_data *q) +{ + get_random_bytes(&q->bins[slot].perturbation, + sizeof(q->bins[slot].perturbation)); +} + +static void sfb_swap_slot(struct sfb_sched_data *q) +{ + sfb_init_perturbation(q->slot, q); + q->slot ^= 1; + q->double_buffering = false; +} + +/* Non elastic flows are allowed to use part of the bandwidth, expressed + * in "penalty_rate" packets per second, with "penalty_burst" burst + */ +static bool sfb_rate_limit(struct sk_buff *skb, struct sfb_sched_data *q) +{ + if (q->penalty_rate == 0 || q->penalty_burst == 0) + return true; + + if (q->tokens_avail < 1) { + unsigned long age = min(10UL * HZ, jiffies - q->token_time); + + q->tokens_avail = (age * q->penalty_rate) / HZ; + if (q->tokens_avail > q->penalty_burst) + q->tokens_avail = q->penalty_burst; + q->token_time = jiffies; + if (q->tokens_avail < 1) + return true; + } + + q->tokens_avail--; + return false; +} + +static bool sfb_classify(struct sk_buff *skb, struct tcf_proto *fl, + int *qerr, u32 *salt) +{ + struct tcf_result res; + int result; + + result = tcf_classify(skb, fl, &res, false); + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return false; + } +#endif + *salt = TC_H_MIN(res.classid); + return true; + } + return false; +} + +static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + + struct sfb_sched_data *q = qdisc_priv(sch); + struct Qdisc *child = q->qdisc; + struct tcf_proto *fl; + int i; + u32 p_min = ~0; + u32 minqlen = ~0; + u32 r, sfbhash; + u32 slot = q->slot; + int ret = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + + if (unlikely(sch->q.qlen >= q->limit)) { + qdisc_qstats_overlimit(sch); + q->stats.queuedrop++; + goto drop; + } + + if (q->rehash_interval > 0) { + unsigned long limit = q->rehash_time + q->rehash_interval; + + if (unlikely(time_after(jiffies, limit))) { + sfb_swap_slot(q); + q->rehash_time = jiffies; + } else if (unlikely(!q->double_buffering && q->warmup_time > 0 && + time_after(jiffies, limit - q->warmup_time))) { + q->double_buffering = true; + } + } + + fl = rcu_dereference_bh(q->filter_list); + if (fl) { + u32 salt; + + /* If using external classifiers, get result and record it. */ + if (!sfb_classify(skb, fl, &ret, &salt)) + goto other_drop; + sfbhash = siphash_1u32(salt, &q->bins[slot].perturbation); + } else { + sfbhash = skb_get_hash_perturb(skb, &q->bins[slot].perturbation); + } + + + if (!sfbhash) + sfbhash = 1; + sfb_skb_cb(skb)->hashes[slot] = sfbhash; + + for (i = 0; i < SFB_LEVELS; i++) { + u32 hash = sfbhash & SFB_BUCKET_MASK; + struct sfb_bucket *b = &q->bins[slot].bins[i][hash]; + + sfbhash >>= SFB_BUCKET_SHIFT; + if (b->qlen == 0) + decrement_prob(b, q); + else if (b->qlen >= q->bin_size) + increment_prob(b, q); + if (minqlen > b->qlen) + minqlen = b->qlen; + if (p_min > b->p_mark) + p_min = b->p_mark; + } + + slot ^= 1; + sfb_skb_cb(skb)->hashes[slot] = 0; + + if (unlikely(minqlen >= q->max)) { + qdisc_qstats_overlimit(sch); + q->stats.bucketdrop++; + goto drop; + } + + if (unlikely(p_min >= SFB_MAX_PROB)) { + /* Inelastic flow */ + if (q->double_buffering) { + sfbhash = skb_get_hash_perturb(skb, + &q->bins[slot].perturbation); + if (!sfbhash) + sfbhash = 1; + sfb_skb_cb(skb)->hashes[slot] = sfbhash; + + for (i = 0; i < SFB_LEVELS; i++) { + u32 hash = sfbhash & SFB_BUCKET_MASK; + struct sfb_bucket *b = &q->bins[slot].bins[i][hash]; + + sfbhash >>= SFB_BUCKET_SHIFT; + if (b->qlen == 0) + decrement_prob(b, q); + else if (b->qlen >= q->bin_size) + increment_prob(b, q); + } + } + if (sfb_rate_limit(skb, q)) { + qdisc_qstats_overlimit(sch); + q->stats.penaltydrop++; + goto drop; + } + goto enqueue; + } + + r = prandom_u32() & SFB_MAX_PROB; + + if (unlikely(r < p_min)) { + if (unlikely(p_min > SFB_MAX_PROB / 2)) { + /* If we're marking that many packets, then either + * this flow is unresponsive, or we're badly congested. + * In either case, we want to start dropping packets. + */ + if (r < (p_min - SFB_MAX_PROB / 2) * 2) { + q->stats.earlydrop++; + goto drop; + } + } + if (INET_ECN_set_ce(skb)) { + q->stats.marked++; + } else { + q->stats.earlydrop++; + goto drop; + } + } + +enqueue: + ret = qdisc_enqueue(skb, child, to_free); + if (likely(ret == NET_XMIT_SUCCESS)) { + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + increment_qlen(skb, q); + } else if (net_xmit_drop_count(ret)) { + q->stats.childdrop++; + qdisc_qstats_drop(sch); + } + return ret; + +drop: + qdisc_drop(skb, sch, to_free); + return NET_XMIT_CN; +other_drop: + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + kfree_skb(skb); + return ret; +} + +static struct sk_buff *sfb_dequeue(struct Qdisc *sch) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + struct Qdisc *child = q->qdisc; + struct sk_buff *skb; + + skb = child->dequeue(q->qdisc); + + if (skb) { + qdisc_bstats_update(sch, skb); + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + decrement_qlen(skb, q); + } + + return skb; +} + +static struct sk_buff *sfb_peek(struct Qdisc *sch) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + struct Qdisc *child = q->qdisc; + + return child->ops->peek(child); +} + +/* No sfb_drop -- impossible since the child doesn't return the dropped skb. */ + +static void sfb_reset(struct Qdisc *sch) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + + qdisc_reset(q->qdisc); + sch->qstats.backlog = 0; + sch->q.qlen = 0; + q->slot = 0; + q->double_buffering = false; + sfb_zero_all_buckets(q); + sfb_init_perturbation(0, q); +} + +static void sfb_destroy(struct Qdisc *sch) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + + tcf_block_put(q->block); + qdisc_put(q->qdisc); +} + +static const struct nla_policy sfb_policy[TCA_SFB_MAX + 1] = { + [TCA_SFB_PARMS] = { .len = sizeof(struct tc_sfb_qopt) }, +}; + +static const struct tc_sfb_qopt sfb_default_ops = { + .rehash_interval = 600 * MSEC_PER_SEC, + .warmup_time = 60 * MSEC_PER_SEC, + .limit = 0, + .max = 25, + .bin_size = 20, + .increment = (SFB_MAX_PROB + 500) / 1000, /* 0.1 % */ + .decrement = (SFB_MAX_PROB + 3000) / 6000, + .penalty_rate = 10, + .penalty_burst = 20, +}; + +static int sfb_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + struct Qdisc *child; + struct nlattr *tb[TCA_SFB_MAX + 1]; + const struct tc_sfb_qopt *ctl = &sfb_default_ops; + u32 limit; + int err; + + if (opt) { + err = nla_parse_nested(tb, TCA_SFB_MAX, opt, sfb_policy, NULL); + if (err < 0) + return -EINVAL; + + if (tb[TCA_SFB_PARMS] == NULL) + return -EINVAL; + + ctl = nla_data(tb[TCA_SFB_PARMS]); + } + + limit = ctl->limit; + if (limit == 0) + limit = qdisc_dev(sch)->tx_queue_len; + + child = fifo_create_dflt(sch, &pfifo_qdisc_ops, limit, extack); + if (IS_ERR(child)) + return PTR_ERR(child); + + if (child != &noop_qdisc) + qdisc_hash_add(child, true); + sch_tree_lock(sch); + + qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen, + q->qdisc->qstats.backlog); + qdisc_put(q->qdisc); + q->qdisc = child; + + q->rehash_interval = msecs_to_jiffies(ctl->rehash_interval); + q->warmup_time = msecs_to_jiffies(ctl->warmup_time); + q->rehash_time = jiffies; + q->limit = limit; + q->increment = ctl->increment; + q->decrement = ctl->decrement; + q->max = ctl->max; + q->bin_size = ctl->bin_size; + q->penalty_rate = ctl->penalty_rate; + q->penalty_burst = ctl->penalty_burst; + q->tokens_avail = ctl->penalty_burst; + q->token_time = jiffies; + + q->slot = 0; + q->double_buffering = false; + sfb_zero_all_buckets(q); + sfb_init_perturbation(0, q); + sfb_init_perturbation(1, q); + + sch_tree_unlock(sch); + + return 0; +} + +static int sfb_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + int err; + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + q->qdisc = &noop_qdisc; + return sfb_change(sch, opt, extack); +} + +static int sfb_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + struct nlattr *opts; + struct tc_sfb_qopt opt = { + .rehash_interval = jiffies_to_msecs(q->rehash_interval), + .warmup_time = jiffies_to_msecs(q->warmup_time), + .limit = q->limit, + .max = q->max, + .bin_size = q->bin_size, + .increment = q->increment, + .decrement = q->decrement, + .penalty_rate = q->penalty_rate, + .penalty_burst = q->penalty_burst, + }; + + sch->qstats.backlog = q->qdisc->qstats.backlog; + opts = nla_nest_start(skb, TCA_OPTIONS); + if (opts == NULL) + goto nla_put_failure; + if (nla_put(skb, TCA_SFB_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + return nla_nest_end(skb, opts); + +nla_put_failure: + nla_nest_cancel(skb, opts); + return -EMSGSIZE; +} + +static int sfb_dump_stats(struct Qdisc *sch, struct gnet_dump *d) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + struct tc_sfb_xstats st = { + .earlydrop = q->stats.earlydrop, + .penaltydrop = q->stats.penaltydrop, + .bucketdrop = q->stats.bucketdrop, + .queuedrop = q->stats.queuedrop, + .childdrop = q->stats.childdrop, + .marked = q->stats.marked, + }; + + st.maxqlen = sfb_compute_qlen(&st.maxprob, &st.avgprob, q); + + return gnet_stats_copy_app(d, &st, sizeof(st)); +} + +static int sfb_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + return -ENOSYS; +} + +static int sfb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + + if (new == NULL) + new = &noop_qdisc; + + *old = qdisc_replace(sch, new, &q->qdisc); + return 0; +} + +static struct Qdisc *sfb_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + + return q->qdisc; +} + +static unsigned long sfb_find(struct Qdisc *sch, u32 classid) +{ + return 1; +} + +static void sfb_unbind(struct Qdisc *sch, unsigned long arg) +{ +} + +static int sfb_change_class(struct Qdisc *sch, u32 classid, u32 parentid, + struct nlattr **tca, unsigned long *arg, + struct netlink_ext_ack *extack) +{ + return -ENOSYS; +} + +static int sfb_delete(struct Qdisc *sch, unsigned long cl) +{ + return -ENOSYS; +} + +static void sfb_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + if (!walker->stop) { + if (walker->count >= walker->skip) + if (walker->fn(sch, 1, walker) < 0) { + walker->stop = 1; + return; + } + walker->count++; + } +} + +static struct tcf_block *sfb_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct sfb_sched_data *q = qdisc_priv(sch); + + if (cl) + return NULL; + return q->block; +} + +static unsigned long sfb_bind(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + return 0; +} + + +static const struct Qdisc_class_ops sfb_class_ops = { + .graft = sfb_graft, + .leaf = sfb_leaf, + .find = sfb_find, + .change = sfb_change_class, + .delete = sfb_delete, + .walk = sfb_walk, + .tcf_block = sfb_tcf_block, + .bind_tcf = sfb_bind, + .unbind_tcf = sfb_unbind, + .dump = sfb_dump_class, +}; + +static struct Qdisc_ops sfb_qdisc_ops __read_mostly = { + .id = "sfb", + .priv_size = sizeof(struct sfb_sched_data), + .cl_ops = &sfb_class_ops, + .enqueue = sfb_enqueue, + .dequeue = sfb_dequeue, + .peek = sfb_peek, + .init = sfb_init, + .reset = sfb_reset, + .destroy = sfb_destroy, + .change = sfb_change, + .dump = sfb_dump, + .dump_stats = sfb_dump_stats, + .owner = THIS_MODULE, +}; + +static int __init sfb_module_init(void) +{ + return register_qdisc(&sfb_qdisc_ops); +} + +static void __exit sfb_module_exit(void) +{ + unregister_qdisc(&sfb_qdisc_ops); +} + +module_init(sfb_module_init) +module_exit(sfb_module_exit) + +MODULE_DESCRIPTION("Stochastic Fair Blue queue discipline"); +MODULE_AUTHOR("Juliusz Chroboczek"); +MODULE_AUTHOR("Eric Dumazet"); +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c new file mode 100644 index 000000000..1bfdf90fa --- /dev/null +++ b/net/sched/sch_sfq.c @@ -0,0 +1,944 @@ +/* + * net/sched/sch_sfq.c Stochastic Fairness Queueing discipline. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/jiffies.h> +#include <linux/string.h> +#include <linux/in.h> +#include <linux/errno.h> +#include <linux/init.h> +#include <linux/skbuff.h> +#include <linux/siphash.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <net/pkt_cls.h> +#include <net/red.h> + + +/* Stochastic Fairness Queuing algorithm. + ======================================= + + Source: + Paul E. McKenney "Stochastic Fairness Queuing", + IEEE INFOCOMM'90 Proceedings, San Francisco, 1990. + + Paul E. McKenney "Stochastic Fairness Queuing", + "Interworking: Research and Experience", v.2, 1991, p.113-131. + + + See also: + M. Shreedhar and George Varghese "Efficient Fair + Queuing using Deficit Round Robin", Proc. SIGCOMM 95. + + + This is not the thing that is usually called (W)FQ nowadays. + It does not use any timestamp mechanism, but instead + processes queues in round-robin order. + + ADVANTAGE: + + - It is very cheap. Both CPU and memory requirements are minimal. + + DRAWBACKS: + + - "Stochastic" -> It is not 100% fair. + When hash collisions occur, several flows are considered as one. + + - "Round-robin" -> It introduces larger delays than virtual clock + based schemes, and should not be used for isolating interactive + traffic from non-interactive. It means, that this scheduler + should be used as leaf of CBQ or P3, which put interactive traffic + to higher priority band. + + We still need true WFQ for top level CSZ, but using WFQ + for the best effort traffic is absolutely pointless: + SFQ is superior for this purpose. + + IMPLEMENTATION: + This implementation limits : + - maximal queue length per flow to 127 packets. + - max mtu to 2^18-1; + - max 65408 flows, + - number of hash buckets to 65536. + + It is easy to increase these values, but not in flight. */ + +#define SFQ_MAX_DEPTH 127 /* max number of packets per flow */ +#define SFQ_DEFAULT_FLOWS 128 +#define SFQ_MAX_FLOWS (0x10000 - SFQ_MAX_DEPTH - 1) /* max number of flows */ +#define SFQ_EMPTY_SLOT 0xffff +#define SFQ_DEFAULT_HASH_DIVISOR 1024 + +/* We use 16 bits to store allot, and want to handle packets up to 64K + * Scale allot by 8 (1<<3) so that no overflow occurs. + */ +#define SFQ_ALLOT_SHIFT 3 +#define SFQ_ALLOT_SIZE(X) DIV_ROUND_UP(X, 1 << SFQ_ALLOT_SHIFT) + +/* This type should contain at least SFQ_MAX_DEPTH + 1 + SFQ_MAX_FLOWS values */ +typedef u16 sfq_index; + +/* + * We dont use pointers to save space. + * Small indexes [0 ... SFQ_MAX_FLOWS - 1] are 'pointers' to slots[] array + * while following values [SFQ_MAX_FLOWS ... SFQ_MAX_FLOWS + SFQ_MAX_DEPTH] + * are 'pointers' to dep[] array + */ +struct sfq_head { + sfq_index next; + sfq_index prev; +}; + +struct sfq_slot { + struct sk_buff *skblist_next; + struct sk_buff *skblist_prev; + sfq_index qlen; /* number of skbs in skblist */ + sfq_index next; /* next slot in sfq RR chain */ + struct sfq_head dep; /* anchor in dep[] chains */ + unsigned short hash; /* hash value (index in ht[]) */ + short allot; /* credit for this slot */ + + unsigned int backlog; + struct red_vars vars; +}; + +struct sfq_sched_data { +/* frequently used fields */ + int limit; /* limit of total number of packets in this qdisc */ + unsigned int divisor; /* number of slots in hash table */ + u8 headdrop; + u8 maxdepth; /* limit of packets per flow */ + + siphash_key_t perturbation; + u8 cur_depth; /* depth of longest slot */ + u8 flags; + unsigned short scaled_quantum; /* SFQ_ALLOT_SIZE(quantum) */ + struct tcf_proto __rcu *filter_list; + struct tcf_block *block; + sfq_index *ht; /* Hash table ('divisor' slots) */ + struct sfq_slot *slots; /* Flows table ('maxflows' entries) */ + + struct red_parms *red_parms; + struct tc_sfqred_stats stats; + struct sfq_slot *tail; /* current slot in round */ + + struct sfq_head dep[SFQ_MAX_DEPTH + 1]; + /* Linked lists of slots, indexed by depth + * dep[0] : list of unused flows + * dep[1] : list of flows with 1 packet + * dep[X] : list of flows with X packets + */ + + unsigned int maxflows; /* number of flows in flows array */ + int perturb_period; + unsigned int quantum; /* Allotment per round: MUST BE >= MTU */ + struct timer_list perturb_timer; + struct Qdisc *sch; +}; + +/* + * sfq_head are either in a sfq_slot or in dep[] array + */ +static inline struct sfq_head *sfq_dep_head(struct sfq_sched_data *q, sfq_index val) +{ + if (val < SFQ_MAX_FLOWS) + return &q->slots[val].dep; + return &q->dep[val - SFQ_MAX_FLOWS]; +} + +static unsigned int sfq_hash(const struct sfq_sched_data *q, + const struct sk_buff *skb) +{ + return skb_get_hash_perturb(skb, &q->perturbation) & (q->divisor - 1); +} + +static unsigned int sfq_classify(struct sk_buff *skb, struct Qdisc *sch, + int *qerr) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + struct tcf_result res; + struct tcf_proto *fl; + int result; + + if (TC_H_MAJ(skb->priority) == sch->handle && + TC_H_MIN(skb->priority) > 0 && + TC_H_MIN(skb->priority) <= q->divisor) + return TC_H_MIN(skb->priority); + + fl = rcu_dereference_bh(q->filter_list); + if (!fl) + return sfq_hash(q, skb) + 1; + + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; + result = tcf_classify(skb, fl, &res, false); + if (result >= 0) { +#ifdef CONFIG_NET_CLS_ACT + switch (result) { + case TC_ACT_STOLEN: + case TC_ACT_QUEUED: + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ + case TC_ACT_SHOT: + return 0; + } +#endif + if (TC_H_MIN(res.classid) <= q->divisor) + return TC_H_MIN(res.classid); + } + return 0; +} + +/* + * x : slot number [0 .. SFQ_MAX_FLOWS - 1] + */ +static inline void sfq_link(struct sfq_sched_data *q, sfq_index x) +{ + sfq_index p, n; + struct sfq_slot *slot = &q->slots[x]; + int qlen = slot->qlen; + + p = qlen + SFQ_MAX_FLOWS; + n = q->dep[qlen].next; + + slot->dep.next = n; + slot->dep.prev = p; + + q->dep[qlen].next = x; /* sfq_dep_head(q, p)->next = x */ + sfq_dep_head(q, n)->prev = x; +} + +#define sfq_unlink(q, x, n, p) \ + do { \ + n = q->slots[x].dep.next; \ + p = q->slots[x].dep.prev; \ + sfq_dep_head(q, p)->next = n; \ + sfq_dep_head(q, n)->prev = p; \ + } while (0) + + +static inline void sfq_dec(struct sfq_sched_data *q, sfq_index x) +{ + sfq_index p, n; + int d; + + sfq_unlink(q, x, n, p); + + d = q->slots[x].qlen--; + if (n == p && q->cur_depth == d) + q->cur_depth--; + sfq_link(q, x); +} + +static inline void sfq_inc(struct sfq_sched_data *q, sfq_index x) +{ + sfq_index p, n; + int d; + + sfq_unlink(q, x, n, p); + + d = ++q->slots[x].qlen; + if (q->cur_depth < d) + q->cur_depth = d; + sfq_link(q, x); +} + +/* helper functions : might be changed when/if skb use a standard list_head */ + +/* remove one skb from tail of slot queue */ +static inline struct sk_buff *slot_dequeue_tail(struct sfq_slot *slot) +{ + struct sk_buff *skb = slot->skblist_prev; + + slot->skblist_prev = skb->prev; + skb->prev->next = (struct sk_buff *)slot; + skb->next = skb->prev = NULL; + return skb; +} + +/* remove one skb from head of slot queue */ +static inline struct sk_buff *slot_dequeue_head(struct sfq_slot *slot) +{ + struct sk_buff *skb = slot->skblist_next; + + slot->skblist_next = skb->next; + skb->next->prev = (struct sk_buff *)slot; + skb->next = skb->prev = NULL; + return skb; +} + +static inline void slot_queue_init(struct sfq_slot *slot) +{ + memset(slot, 0, sizeof(*slot)); + slot->skblist_prev = slot->skblist_next = (struct sk_buff *)slot; +} + +/* add skb to slot queue (tail add) */ +static inline void slot_queue_add(struct sfq_slot *slot, struct sk_buff *skb) +{ + skb->prev = slot->skblist_prev; + skb->next = (struct sk_buff *)slot; + slot->skblist_prev->next = skb; + slot->skblist_prev = skb; +} + +static unsigned int sfq_drop(struct Qdisc *sch, struct sk_buff **to_free) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + sfq_index x, d = q->cur_depth; + struct sk_buff *skb; + unsigned int len; + struct sfq_slot *slot; + + /* Queue is full! Find the longest slot and drop tail packet from it */ + if (d > 1) { + x = q->dep[d].next; + slot = &q->slots[x]; +drop: + skb = q->headdrop ? slot_dequeue_head(slot) : slot_dequeue_tail(slot); + len = qdisc_pkt_len(skb); + slot->backlog -= len; + sfq_dec(q, x); + sch->q.qlen--; + qdisc_qstats_backlog_dec(sch, skb); + qdisc_drop(skb, sch, to_free); + return len; + } + + if (d == 1) { + /* It is difficult to believe, but ALL THE SLOTS HAVE LENGTH 1. */ + x = q->tail->next; + slot = &q->slots[x]; + q->tail->next = slot->next; + q->ht[slot->hash] = SFQ_EMPTY_SLOT; + goto drop; + } + + return 0; +} + +/* Is ECN parameter configured */ +static int sfq_prob_mark(const struct sfq_sched_data *q) +{ + return q->flags & TC_RED_ECN; +} + +/* Should packets over max threshold just be marked */ +static int sfq_hard_mark(const struct sfq_sched_data *q) +{ + return (q->flags & (TC_RED_ECN | TC_RED_HARDDROP)) == TC_RED_ECN; +} + +static int sfq_headdrop(const struct sfq_sched_data *q) +{ + return q->headdrop; +} + +static int +sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + unsigned int hash, dropped; + sfq_index x, qlen; + struct sfq_slot *slot; + int uninitialized_var(ret); + struct sk_buff *head; + int delta; + + hash = sfq_classify(skb, sch, &ret); + if (hash == 0) { + if (ret & __NET_XMIT_BYPASS) + qdisc_qstats_drop(sch); + __qdisc_drop(skb, to_free); + return ret; + } + hash--; + + x = q->ht[hash]; + slot = &q->slots[x]; + if (x == SFQ_EMPTY_SLOT) { + x = q->dep[0].next; /* get a free slot */ + if (x >= SFQ_MAX_FLOWS) + return qdisc_drop(skb, sch, to_free); + q->ht[hash] = x; + slot = &q->slots[x]; + slot->hash = hash; + slot->backlog = 0; /* should already be 0 anyway... */ + red_set_vars(&slot->vars); + goto enqueue; + } + if (q->red_parms) { + slot->vars.qavg = red_calc_qavg_no_idle_time(q->red_parms, + &slot->vars, + slot->backlog); + switch (red_action(q->red_parms, + &slot->vars, + slot->vars.qavg)) { + case RED_DONT_MARK: + break; + + case RED_PROB_MARK: + qdisc_qstats_overlimit(sch); + if (sfq_prob_mark(q)) { + /* We know we have at least one packet in queue */ + if (sfq_headdrop(q) && + INET_ECN_set_ce(slot->skblist_next)) { + q->stats.prob_mark_head++; + break; + } + if (INET_ECN_set_ce(skb)) { + q->stats.prob_mark++; + break; + } + } + q->stats.prob_drop++; + goto congestion_drop; + + case RED_HARD_MARK: + qdisc_qstats_overlimit(sch); + if (sfq_hard_mark(q)) { + /* We know we have at least one packet in queue */ + if (sfq_headdrop(q) && + INET_ECN_set_ce(slot->skblist_next)) { + q->stats.forced_mark_head++; + break; + } + if (INET_ECN_set_ce(skb)) { + q->stats.forced_mark++; + break; + } + } + q->stats.forced_drop++; + goto congestion_drop; + } + } + + if (slot->qlen >= q->maxdepth) { +congestion_drop: + if (!sfq_headdrop(q)) + return qdisc_drop(skb, sch, to_free); + + /* We know we have at least one packet in queue */ + head = slot_dequeue_head(slot); + delta = qdisc_pkt_len(head) - qdisc_pkt_len(skb); + sch->qstats.backlog -= delta; + slot->backlog -= delta; + qdisc_drop(head, sch, to_free); + + slot_queue_add(slot, skb); + qdisc_tree_reduce_backlog(sch, 0, delta); + return NET_XMIT_CN; + } + +enqueue: + qdisc_qstats_backlog_inc(sch, skb); + slot->backlog += qdisc_pkt_len(skb); + slot_queue_add(slot, skb); + sfq_inc(q, x); + if (slot->qlen == 1) { /* The flow is new */ + if (q->tail == NULL) { /* It is the first flow */ + slot->next = x; + } else { + slot->next = q->tail->next; + q->tail->next = x; + } + /* We put this flow at the end of our flow list. + * This might sound unfair for a new flow to wait after old ones, + * but we could endup servicing new flows only, and freeze old ones. + */ + q->tail = slot; + /* We could use a bigger initial quantum for new flows */ + slot->allot = q->scaled_quantum; + } + if (++sch->q.qlen <= q->limit) + return NET_XMIT_SUCCESS; + + qlen = slot->qlen; + dropped = sfq_drop(sch, to_free); + /* Return Congestion Notification only if we dropped a packet + * from this flow. + */ + if (qlen != slot->qlen) { + qdisc_tree_reduce_backlog(sch, 0, dropped - qdisc_pkt_len(skb)); + return NET_XMIT_CN; + } + + /* As we dropped a packet, better let upper stack know this */ + qdisc_tree_reduce_backlog(sch, 1, dropped); + return NET_XMIT_SUCCESS; +} + +static struct sk_buff * +sfq_dequeue(struct Qdisc *sch) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + sfq_index a, next_a; + struct sfq_slot *slot; + + /* No active slots */ + if (q->tail == NULL) + return NULL; + +next_slot: + a = q->tail->next; + slot = &q->slots[a]; + if (slot->allot <= 0) { + q->tail = slot; + slot->allot += q->scaled_quantum; + goto next_slot; + } + skb = slot_dequeue_head(slot); + sfq_dec(q, a); + qdisc_bstats_update(sch, skb); + sch->q.qlen--; + qdisc_qstats_backlog_dec(sch, skb); + slot->backlog -= qdisc_pkt_len(skb); + /* Is the slot empty? */ + if (slot->qlen == 0) { + q->ht[slot->hash] = SFQ_EMPTY_SLOT; + next_a = slot->next; + if (a == next_a) { + q->tail = NULL; /* no more active slots */ + return skb; + } + q->tail->next = next_a; + } else { + slot->allot -= SFQ_ALLOT_SIZE(qdisc_pkt_len(skb)); + } + return skb; +} + +static void +sfq_reset(struct Qdisc *sch) +{ + struct sk_buff *skb; + + while ((skb = sfq_dequeue(sch)) != NULL) + rtnl_kfree_skbs(skb, skb); +} + +/* + * When q->perturbation is changed, we rehash all queued skbs + * to avoid OOO (Out Of Order) effects. + * We dont use sfq_dequeue()/sfq_enqueue() because we dont want to change + * counters. + */ +static void sfq_rehash(struct Qdisc *sch) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + int i; + struct sfq_slot *slot; + struct sk_buff_head list; + int dropped = 0; + unsigned int drop_len = 0; + + __skb_queue_head_init(&list); + + for (i = 0; i < q->maxflows; i++) { + slot = &q->slots[i]; + if (!slot->qlen) + continue; + while (slot->qlen) { + skb = slot_dequeue_head(slot); + sfq_dec(q, i); + __skb_queue_tail(&list, skb); + } + slot->backlog = 0; + red_set_vars(&slot->vars); + q->ht[slot->hash] = SFQ_EMPTY_SLOT; + } + q->tail = NULL; + + while ((skb = __skb_dequeue(&list)) != NULL) { + unsigned int hash = sfq_hash(q, skb); + sfq_index x = q->ht[hash]; + + slot = &q->slots[x]; + if (x == SFQ_EMPTY_SLOT) { + x = q->dep[0].next; /* get a free slot */ + if (x >= SFQ_MAX_FLOWS) { +drop: + qdisc_qstats_backlog_dec(sch, skb); + drop_len += qdisc_pkt_len(skb); + kfree_skb(skb); + dropped++; + continue; + } + q->ht[hash] = x; + slot = &q->slots[x]; + slot->hash = hash; + } + if (slot->qlen >= q->maxdepth) + goto drop; + slot_queue_add(slot, skb); + if (q->red_parms) + slot->vars.qavg = red_calc_qavg(q->red_parms, + &slot->vars, + slot->backlog); + slot->backlog += qdisc_pkt_len(skb); + sfq_inc(q, x); + if (slot->qlen == 1) { /* The flow is new */ + if (q->tail == NULL) { /* It is the first flow */ + slot->next = x; + } else { + slot->next = q->tail->next; + q->tail->next = x; + } + q->tail = slot; + slot->allot = q->scaled_quantum; + } + } + sch->q.qlen -= dropped; + qdisc_tree_reduce_backlog(sch, dropped, drop_len); +} + +static void sfq_perturbation(struct timer_list *t) +{ + struct sfq_sched_data *q = from_timer(q, t, perturb_timer); + struct Qdisc *sch = q->sch; + spinlock_t *root_lock = qdisc_lock(qdisc_root_sleeping(sch)); + siphash_key_t nkey; + + get_random_bytes(&nkey, sizeof(nkey)); + spin_lock(root_lock); + q->perturbation = nkey; + if (!q->filter_list && q->tail) + sfq_rehash(sch); + spin_unlock(root_lock); + + if (q->perturb_period) + mod_timer(&q->perturb_timer, jiffies + q->perturb_period); +} + +static int sfq_change(struct Qdisc *sch, struct nlattr *opt) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + struct tc_sfq_qopt *ctl = nla_data(opt); + struct tc_sfq_qopt_v1 *ctl_v1 = NULL; + unsigned int qlen, dropped = 0; + struct red_parms *p = NULL; + struct sk_buff *to_free = NULL; + struct sk_buff *tail = NULL; + + if (opt->nla_len < nla_attr_size(sizeof(*ctl))) + return -EINVAL; + if (opt->nla_len >= nla_attr_size(sizeof(*ctl_v1))) + ctl_v1 = nla_data(opt); + if (ctl->divisor && + (!is_power_of_2(ctl->divisor) || ctl->divisor > 65536)) + return -EINVAL; + + /* slot->allot is a short, make sure quantum is not too big. */ + if (ctl->quantum) { + unsigned int scaled = SFQ_ALLOT_SIZE(ctl->quantum); + + if (scaled <= 0 || scaled > SHRT_MAX) + return -EINVAL; + } + + if (ctl_v1 && !red_check_params(ctl_v1->qth_min, ctl_v1->qth_max, + ctl_v1->Wlog, ctl_v1->Scell_log, NULL)) + return -EINVAL; + if (ctl_v1 && ctl_v1->qth_min) { + p = kmalloc(sizeof(*p), GFP_KERNEL); + if (!p) + return -ENOMEM; + } + sch_tree_lock(sch); + if (ctl->quantum) { + q->quantum = ctl->quantum; + q->scaled_quantum = SFQ_ALLOT_SIZE(q->quantum); + } + q->perturb_period = ctl->perturb_period * HZ; + if (ctl->flows) + q->maxflows = min_t(u32, ctl->flows, SFQ_MAX_FLOWS); + if (ctl->divisor) { + q->divisor = ctl->divisor; + q->maxflows = min_t(u32, q->maxflows, q->divisor); + } + if (ctl_v1) { + if (ctl_v1->depth) + q->maxdepth = min_t(u32, ctl_v1->depth, SFQ_MAX_DEPTH); + if (p) { + swap(q->red_parms, p); + red_set_parms(q->red_parms, + ctl_v1->qth_min, ctl_v1->qth_max, + ctl_v1->Wlog, + ctl_v1->Plog, ctl_v1->Scell_log, + NULL, + ctl_v1->max_P); + } + q->flags = ctl_v1->flags; + q->headdrop = ctl_v1->headdrop; + } + if (ctl->limit) { + q->limit = min_t(u32, ctl->limit, q->maxdepth * q->maxflows); + q->maxflows = min_t(u32, q->maxflows, q->limit); + } + + qlen = sch->q.qlen; + while (sch->q.qlen > q->limit) { + dropped += sfq_drop(sch, &to_free); + if (!tail) + tail = to_free; + } + + rtnl_kfree_skbs(to_free, tail); + qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped); + + del_timer(&q->perturb_timer); + if (q->perturb_period) { + mod_timer(&q->perturb_timer, jiffies + q->perturb_period); + get_random_bytes(&q->perturbation, sizeof(q->perturbation)); + } + sch_tree_unlock(sch); + kfree(p); + return 0; +} + +static void *sfq_alloc(size_t sz) +{ + return kvmalloc(sz, GFP_KERNEL); +} + +static void sfq_free(void *addr) +{ + kvfree(addr); +} + +static void sfq_destroy(struct Qdisc *sch) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + + tcf_block_put(q->block); + q->perturb_period = 0; + del_timer_sync(&q->perturb_timer); + sfq_free(q->ht); + sfq_free(q->slots); + kfree(q->red_parms); +} + +static int sfq_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + int i; + int err; + + q->sch = sch; + timer_setup(&q->perturb_timer, sfq_perturbation, TIMER_DEFERRABLE); + + err = tcf_block_get(&q->block, &q->filter_list, sch, extack); + if (err) + return err; + + for (i = 0; i < SFQ_MAX_DEPTH + 1; i++) { + q->dep[i].next = i + SFQ_MAX_FLOWS; + q->dep[i].prev = i + SFQ_MAX_FLOWS; + } + + q->limit = SFQ_MAX_DEPTH; + q->maxdepth = SFQ_MAX_DEPTH; + q->cur_depth = 0; + q->tail = NULL; + q->divisor = SFQ_DEFAULT_HASH_DIVISOR; + q->maxflows = SFQ_DEFAULT_FLOWS; + q->quantum = psched_mtu(qdisc_dev(sch)); + q->scaled_quantum = SFQ_ALLOT_SIZE(q->quantum); + q->perturb_period = 0; + get_random_bytes(&q->perturbation, sizeof(q->perturbation)); + + if (opt) { + int err = sfq_change(sch, opt); + if (err) + return err; + } + + q->ht = sfq_alloc(sizeof(q->ht[0]) * q->divisor); + q->slots = sfq_alloc(sizeof(q->slots[0]) * q->maxflows); + if (!q->ht || !q->slots) { + /* Note: sfq_destroy() will be called by our caller */ + return -ENOMEM; + } + + for (i = 0; i < q->divisor; i++) + q->ht[i] = SFQ_EMPTY_SLOT; + + for (i = 0; i < q->maxflows; i++) { + slot_queue_init(&q->slots[i]); + sfq_link(q, i); + } + if (q->limit >= 1) + sch->flags |= TCQ_F_CAN_BYPASS; + else + sch->flags &= ~TCQ_F_CAN_BYPASS; + return 0; +} + +static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + unsigned char *b = skb_tail_pointer(skb); + struct tc_sfq_qopt_v1 opt; + struct red_parms *p = q->red_parms; + + memset(&opt, 0, sizeof(opt)); + opt.v0.quantum = q->quantum; + opt.v0.perturb_period = q->perturb_period / HZ; + opt.v0.limit = q->limit; + opt.v0.divisor = q->divisor; + opt.v0.flows = q->maxflows; + opt.depth = q->maxdepth; + opt.headdrop = q->headdrop; + + if (p) { + opt.qth_min = p->qth_min >> p->Wlog; + opt.qth_max = p->qth_max >> p->Wlog; + opt.Wlog = p->Wlog; + opt.Plog = p->Plog; + opt.Scell_log = p->Scell_log; + opt.max_P = p->max_P; + } + memcpy(&opt.stats, &q->stats, sizeof(opt.stats)); + opt.flags = q->flags; + + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + return -1; +} + +static struct Qdisc *sfq_leaf(struct Qdisc *sch, unsigned long arg) +{ + return NULL; +} + +static unsigned long sfq_find(struct Qdisc *sch, u32 classid) +{ + return 0; +} + +static unsigned long sfq_bind(struct Qdisc *sch, unsigned long parent, + u32 classid) +{ + return 0; +} + +static void sfq_unbind(struct Qdisc *q, unsigned long cl) +{ +} + +static struct tcf_block *sfq_tcf_block(struct Qdisc *sch, unsigned long cl, + struct netlink_ext_ack *extack) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + + if (cl) + return NULL; + return q->block; +} + +static int sfq_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + tcm->tcm_handle |= TC_H_MIN(cl); + return 0; +} + +static int sfq_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + sfq_index idx = q->ht[cl - 1]; + struct gnet_stats_queue qs = { 0 }; + struct tc_sfq_xstats xstats = { 0 }; + + if (idx != SFQ_EMPTY_SLOT) { + const struct sfq_slot *slot = &q->slots[idx]; + + xstats.allot = slot->allot << SFQ_ALLOT_SHIFT; + qs.qlen = slot->qlen; + qs.backlog = slot->backlog; + } + if (gnet_stats_copy_queue(d, NULL, &qs, qs.qlen) < 0) + return -1; + return gnet_stats_copy_app(d, &xstats, sizeof(xstats)); +} + +static void sfq_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + struct sfq_sched_data *q = qdisc_priv(sch); + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < q->divisor; i++) { + if (q->ht[i] == SFQ_EMPTY_SLOT || + arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, i + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } +} + +static const struct Qdisc_class_ops sfq_class_ops = { + .leaf = sfq_leaf, + .find = sfq_find, + .tcf_block = sfq_tcf_block, + .bind_tcf = sfq_bind, + .unbind_tcf = sfq_unbind, + .dump = sfq_dump_class, + .dump_stats = sfq_dump_class_stats, + .walk = sfq_walk, +}; + +static struct Qdisc_ops sfq_qdisc_ops __read_mostly = { + .cl_ops = &sfq_class_ops, + .id = "sfq", + .priv_size = sizeof(struct sfq_sched_data), + .enqueue = sfq_enqueue, + .dequeue = sfq_dequeue, + .peek = qdisc_peek_dequeued, + .init = sfq_init, + .reset = sfq_reset, + .destroy = sfq_destroy, + .change = NULL, + .dump = sfq_dump, + .owner = THIS_MODULE, +}; + +static int __init sfq_module_init(void) +{ + return register_qdisc(&sfq_qdisc_ops); +} +static void __exit sfq_module_exit(void) +{ + unregister_qdisc(&sfq_qdisc_ops); +} +module_init(sfq_module_init) +module_exit(sfq_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_skbprio.c b/net/sched/sch_skbprio.c new file mode 100644 index 000000000..3d9de5284 --- /dev/null +++ b/net/sched/sch_skbprio.c @@ -0,0 +1,323 @@ +/* + * net/sched/sch_skbprio.c SKB Priority Queue. + * + * 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. + * + * Authors: Nishanth Devarajan, <ndev2021@gmail.com> + * Cody Doucette, <doucette@bu.edu> + * original idea by Michel Machado, Cody Doucette, and Qiaobin Fu + */ + +#include <linux/string.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/pkt_sched.h> +#include <net/sch_generic.h> +#include <net/inet_ecn.h> + +/* SKB Priority Queue + * ================================= + * + * Skbprio (SKB Priority Queue) is a queueing discipline that prioritizes + * packets according to their skb->priority field. Under congestion, + * Skbprio drops already-enqueued lower priority packets to make space + * available for higher priority packets; it was conceived as a solution + * for denial-of-service defenses that need to route packets with different + * priorities as a mean to overcome DoS attacks. + */ + +struct skbprio_sched_data { + /* Queue state. */ + struct sk_buff_head qdiscs[SKBPRIO_MAX_PRIORITY]; + struct gnet_stats_queue qstats[SKBPRIO_MAX_PRIORITY]; + u16 highest_prio; + u16 lowest_prio; +}; + +static u16 calc_new_high_prio(const struct skbprio_sched_data *q) +{ + int prio; + + for (prio = q->highest_prio - 1; prio >= q->lowest_prio; prio--) { + if (!skb_queue_empty(&q->qdiscs[prio])) + return prio; + } + + /* SKB queue is empty, return 0 (default highest priority setting). */ + return 0; +} + +static u16 calc_new_low_prio(const struct skbprio_sched_data *q) +{ + int prio; + + for (prio = q->lowest_prio + 1; prio <= q->highest_prio; prio++) { + if (!skb_queue_empty(&q->qdiscs[prio])) + return prio; + } + + /* SKB queue is empty, return SKBPRIO_MAX_PRIORITY - 1 + * (default lowest priority setting). + */ + return SKBPRIO_MAX_PRIORITY - 1; +} + +static int skbprio_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + const unsigned int max_priority = SKBPRIO_MAX_PRIORITY - 1; + struct skbprio_sched_data *q = qdisc_priv(sch); + struct sk_buff_head *qdisc; + struct sk_buff_head *lp_qdisc; + struct sk_buff *to_drop; + u16 prio, lp; + + /* Obtain the priority of @skb. */ + prio = min(skb->priority, max_priority); + + qdisc = &q->qdiscs[prio]; + if (sch->q.qlen < sch->limit) { + __skb_queue_tail(qdisc, skb); + qdisc_qstats_backlog_inc(sch, skb); + q->qstats[prio].backlog += qdisc_pkt_len(skb); + + /* Check to update highest and lowest priorities. */ + if (prio > q->highest_prio) + q->highest_prio = prio; + + if (prio < q->lowest_prio) + q->lowest_prio = prio; + + sch->q.qlen++; + return NET_XMIT_SUCCESS; + } + + /* If this packet has the lowest priority, drop it. */ + lp = q->lowest_prio; + if (prio <= lp) { + q->qstats[prio].drops++; + q->qstats[prio].overlimits++; + return qdisc_drop(skb, sch, to_free); + } + + __skb_queue_tail(qdisc, skb); + qdisc_qstats_backlog_inc(sch, skb); + q->qstats[prio].backlog += qdisc_pkt_len(skb); + + /* Drop the packet at the tail of the lowest priority qdisc. */ + lp_qdisc = &q->qdiscs[lp]; + to_drop = __skb_dequeue_tail(lp_qdisc); + BUG_ON(!to_drop); + qdisc_qstats_backlog_dec(sch, to_drop); + qdisc_drop(to_drop, sch, to_free); + + q->qstats[lp].backlog -= qdisc_pkt_len(to_drop); + q->qstats[lp].drops++; + q->qstats[lp].overlimits++; + + /* Check to update highest and lowest priorities. */ + if (skb_queue_empty(lp_qdisc)) { + if (q->lowest_prio == q->highest_prio) { + /* The incoming packet is the only packet in queue. */ + BUG_ON(sch->q.qlen != 1); + q->lowest_prio = prio; + q->highest_prio = prio; + } else { + q->lowest_prio = calc_new_low_prio(q); + } + } + + if (prio > q->highest_prio) + q->highest_prio = prio; + + return NET_XMIT_CN; +} + +static struct sk_buff *skbprio_dequeue(struct Qdisc *sch) +{ + struct skbprio_sched_data *q = qdisc_priv(sch); + struct sk_buff_head *hpq = &q->qdiscs[q->highest_prio]; + struct sk_buff *skb = __skb_dequeue(hpq); + + if (unlikely(!skb)) + return NULL; + + sch->q.qlen--; + qdisc_qstats_backlog_dec(sch, skb); + qdisc_bstats_update(sch, skb); + + q->qstats[q->highest_prio].backlog -= qdisc_pkt_len(skb); + + /* Update highest priority field. */ + if (skb_queue_empty(hpq)) { + if (q->lowest_prio == q->highest_prio) { + BUG_ON(sch->q.qlen); + q->highest_prio = 0; + q->lowest_prio = SKBPRIO_MAX_PRIORITY - 1; + } else { + q->highest_prio = calc_new_high_prio(q); + } + } + return skb; +} + +static int skbprio_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct tc_skbprio_qopt *ctl = nla_data(opt); + + if (opt->nla_len != nla_attr_size(sizeof(*ctl))) + return -EINVAL; + + sch->limit = ctl->limit; + return 0; +} + +static int skbprio_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct skbprio_sched_data *q = qdisc_priv(sch); + int prio; + + /* Initialise all queues, one for each possible priority. */ + for (prio = 0; prio < SKBPRIO_MAX_PRIORITY; prio++) + __skb_queue_head_init(&q->qdiscs[prio]); + + memset(&q->qstats, 0, sizeof(q->qstats)); + q->highest_prio = 0; + q->lowest_prio = SKBPRIO_MAX_PRIORITY - 1; + sch->limit = 64; + if (!opt) + return 0; + + return skbprio_change(sch, opt, extack); +} + +static int skbprio_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct tc_skbprio_qopt opt; + + opt.limit = sch->limit; + + if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + return -1; + + return skb->len; +} + +static void skbprio_reset(struct Qdisc *sch) +{ + struct skbprio_sched_data *q = qdisc_priv(sch); + int prio; + + sch->qstats.backlog = 0; + sch->q.qlen = 0; + + for (prio = 0; prio < SKBPRIO_MAX_PRIORITY; prio++) + __skb_queue_purge(&q->qdiscs[prio]); + + memset(&q->qstats, 0, sizeof(q->qstats)); + q->highest_prio = 0; + q->lowest_prio = SKBPRIO_MAX_PRIORITY - 1; +} + +static void skbprio_destroy(struct Qdisc *sch) +{ + struct skbprio_sched_data *q = qdisc_priv(sch); + int prio; + + for (prio = 0; prio < SKBPRIO_MAX_PRIORITY; prio++) + __skb_queue_purge(&q->qdiscs[prio]); +} + +static struct Qdisc *skbprio_leaf(struct Qdisc *sch, unsigned long arg) +{ + return NULL; +} + +static unsigned long skbprio_find(struct Qdisc *sch, u32 classid) +{ + return 0; +} + +static int skbprio_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + tcm->tcm_handle |= TC_H_MIN(cl); + return 0; +} + +static int skbprio_dump_class_stats(struct Qdisc *sch, unsigned long cl, + struct gnet_dump *d) +{ + struct skbprio_sched_data *q = qdisc_priv(sch); + if (gnet_stats_copy_queue(d, NULL, &q->qstats[cl - 1], + q->qstats[cl - 1].qlen) < 0) + return -1; + return 0; +} + +static void skbprio_walk(struct Qdisc *sch, struct qdisc_walker *arg) +{ + unsigned int i; + + if (arg->stop) + return; + + for (i = 0; i < SKBPRIO_MAX_PRIORITY; i++) { + if (arg->count < arg->skip) { + arg->count++; + continue; + } + if (arg->fn(sch, i + 1, arg) < 0) { + arg->stop = 1; + break; + } + arg->count++; + } +} + +static const struct Qdisc_class_ops skbprio_class_ops = { + .leaf = skbprio_leaf, + .find = skbprio_find, + .dump = skbprio_dump_class, + .dump_stats = skbprio_dump_class_stats, + .walk = skbprio_walk, +}; + +static struct Qdisc_ops skbprio_qdisc_ops __read_mostly = { + .cl_ops = &skbprio_class_ops, + .id = "skbprio", + .priv_size = sizeof(struct skbprio_sched_data), + .enqueue = skbprio_enqueue, + .dequeue = skbprio_dequeue, + .peek = qdisc_peek_dequeued, + .init = skbprio_init, + .reset = skbprio_reset, + .change = skbprio_change, + .dump = skbprio_dump, + .destroy = skbprio_destroy, + .owner = THIS_MODULE, +}; + +static int __init skbprio_module_init(void) +{ + return register_qdisc(&skbprio_qdisc_ops); +} + +static void __exit skbprio_module_exit(void) +{ + unregister_qdisc(&skbprio_qdisc_ops); +} + +module_init(skbprio_module_init) +module_exit(skbprio_module_exit) + +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c new file mode 100644 index 000000000..dd29de141 --- /dev/null +++ b/net/sched/sch_tbf.c @@ -0,0 +1,563 @@ +/* + * net/sched/sch_tbf.c Token Bucket Filter queue. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + * Dmitry Torokhov <dtor@mail.ru> - allow attaching inner qdiscs - + * original idea by Martin Devera + * + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <net/netlink.h> +#include <net/sch_generic.h> +#include <net/pkt_sched.h> + + +/* Simple Token Bucket Filter. + ======================================= + + SOURCE. + ------- + + None. + + Description. + ------------ + + A data flow obeys TBF with rate R and depth B, if for any + time interval t_i...t_f the number of transmitted bits + does not exceed B + R*(t_f-t_i). + + Packetized version of this definition: + The sequence of packets of sizes s_i served at moments t_i + obeys TBF, if for any i<=k: + + s_i+....+s_k <= B + R*(t_k - t_i) + + Algorithm. + ---------- + + Let N(t_i) be B/R initially and N(t) grow continuously with time as: + + N(t+delta) = min{B/R, N(t) + delta} + + If the first packet in queue has length S, it may be + transmitted only at the time t_* when S/R <= N(t_*), + and in this case N(t) jumps: + + N(t_* + 0) = N(t_* - 0) - S/R. + + + + Actually, QoS requires two TBF to be applied to a data stream. + One of them controls steady state burst size, another + one with rate P (peak rate) and depth M (equal to link MTU) + limits bursts at a smaller time scale. + + It is easy to see that P>R, and B>M. If P is infinity, this double + TBF is equivalent to a single one. + + When TBF works in reshaping mode, latency is estimated as: + + lat = max ((L-B)/R, (L-M)/P) + + + NOTES. + ------ + + If TBF throttles, it starts a watchdog timer, which will wake it up + when it is ready to transmit. + Note that the minimal timer resolution is 1/HZ. + If no new packets arrive during this period, + or if the device is not awaken by EOI for some previous packet, + TBF can stop its activity for 1/HZ. + + + This means, that with depth B, the maximal rate is + + R_crit = B*HZ + + F.e. for 10Mbit ethernet and HZ=100 the minimal allowed B is ~10Kbytes. + + Note that the peak rate TBF is much more tough: with MTU 1500 + P_crit = 150Kbytes/sec. So, if you need greater peak + rates, use alpha with HZ=1000 :-) + + With classful TBF, limit is just kept for backwards compatibility. + It is passed to the default bfifo qdisc - if the inner qdisc is + changed the limit is not effective anymore. +*/ + +struct tbf_sched_data { +/* Parameters */ + u32 limit; /* Maximal length of backlog: bytes */ + u32 max_size; + s64 buffer; /* Token bucket depth/rate: MUST BE >= MTU/B */ + s64 mtu; + struct psched_ratecfg rate; + struct psched_ratecfg peak; + +/* Variables */ + s64 tokens; /* Current number of B tokens */ + s64 ptokens; /* Current number of P tokens */ + s64 t_c; /* Time check-point */ + struct Qdisc *qdisc; /* Inner qdisc, default - bfifo queue */ + struct qdisc_watchdog watchdog; /* Watchdog timer */ +}; + + +/* Time to Length, convert time in ns to length in bytes + * to determinate how many bytes can be sent in given time. + */ +static u64 psched_ns_t2l(const struct psched_ratecfg *r, + u64 time_in_ns) +{ + /* The formula is : + * len = (time_in_ns * r->rate_bytes_ps) / NSEC_PER_SEC + */ + u64 len = time_in_ns * r->rate_bytes_ps; + + do_div(len, NSEC_PER_SEC); + + if (unlikely(r->linklayer == TC_LINKLAYER_ATM)) { + do_div(len, 53); + len = len * 48; + } + + if (len > r->overhead) + len -= r->overhead; + else + len = 0; + + return len; +} + +/* GSO packet is too big, segment it so that tbf can transmit + * each segment in time + */ +static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + struct sk_buff *segs, *nskb; + netdev_features_t features = netif_skb_features(skb); + unsigned int len = 0, prev_len = qdisc_pkt_len(skb); + int ret, nb; + + segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); + + if (IS_ERR_OR_NULL(segs)) + return qdisc_drop(skb, sch, to_free); + + nb = 0; + while (segs) { + nskb = segs->next; + segs->next = NULL; + qdisc_skb_cb(segs)->pkt_len = segs->len; + len += segs->len; + ret = qdisc_enqueue(segs, q->qdisc, to_free); + if (ret != NET_XMIT_SUCCESS) { + if (net_xmit_drop_count(ret)) + qdisc_qstats_drop(sch); + } else { + nb++; + } + segs = nskb; + } + sch->q.qlen += nb; + if (nb > 1) + qdisc_tree_reduce_backlog(sch, 1 - nb, prev_len - len); + consume_skb(skb); + return nb > 0 ? NET_XMIT_SUCCESS : NET_XMIT_DROP; +} + +static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + int ret; + + if (qdisc_pkt_len(skb) > q->max_size) { + if (skb_is_gso(skb) && + skb_gso_validate_mac_len(skb, q->max_size)) + return tbf_segment(skb, sch, to_free); + return qdisc_drop(skb, sch, to_free); + } + ret = qdisc_enqueue(skb, q->qdisc, to_free); + if (ret != NET_XMIT_SUCCESS) { + if (net_xmit_drop_count(ret)) + qdisc_qstats_drop(sch); + return ret; + } + + qdisc_qstats_backlog_inc(sch, skb); + sch->q.qlen++; + return NET_XMIT_SUCCESS; +} + +static bool tbf_peak_present(const struct tbf_sched_data *q) +{ + return q->peak.rate_bytes_ps; +} + +static struct sk_buff *tbf_dequeue(struct Qdisc *sch) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + struct sk_buff *skb; + + skb = q->qdisc->ops->peek(q->qdisc); + + if (skb) { + s64 now; + s64 toks; + s64 ptoks = 0; + unsigned int len = qdisc_pkt_len(skb); + + now = ktime_get_ns(); + toks = min_t(s64, now - q->t_c, q->buffer); + + if (tbf_peak_present(q)) { + ptoks = toks + q->ptokens; + if (ptoks > q->mtu) + ptoks = q->mtu; + ptoks -= (s64) psched_l2t_ns(&q->peak, len); + } + toks += q->tokens; + if (toks > q->buffer) + toks = q->buffer; + toks -= (s64) psched_l2t_ns(&q->rate, len); + + if ((toks|ptoks) >= 0) { + skb = qdisc_dequeue_peeked(q->qdisc); + if (unlikely(!skb)) + return NULL; + + q->t_c = now; + q->tokens = toks; + q->ptokens = ptoks; + qdisc_qstats_backlog_dec(sch, skb); + sch->q.qlen--; + qdisc_bstats_update(sch, skb); + return skb; + } + + qdisc_watchdog_schedule_ns(&q->watchdog, + now + max_t(long, -toks, -ptoks)); + + /* Maybe we have a shorter packet in the queue, + which can be sent now. It sounds cool, + but, however, this is wrong in principle. + We MUST NOT reorder packets under these circumstances. + + Really, if we split the flow into independent + subflows, it would be a very good solution. + This is the main idea of all FQ algorithms + (cf. CSZ, HPFQ, HFSC) + */ + + qdisc_qstats_overlimit(sch); + } + return NULL; +} + +static void tbf_reset(struct Qdisc *sch) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + + qdisc_reset(q->qdisc); + sch->qstats.backlog = 0; + sch->q.qlen = 0; + q->t_c = ktime_get_ns(); + q->tokens = q->buffer; + q->ptokens = q->mtu; + qdisc_watchdog_cancel(&q->watchdog); +} + +static const struct nla_policy tbf_policy[TCA_TBF_MAX + 1] = { + [TCA_TBF_PARMS] = { .len = sizeof(struct tc_tbf_qopt) }, + [TCA_TBF_RTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, + [TCA_TBF_PTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, + [TCA_TBF_RATE64] = { .type = NLA_U64 }, + [TCA_TBF_PRATE64] = { .type = NLA_U64 }, + [TCA_TBF_BURST] = { .type = NLA_U32 }, + [TCA_TBF_PBURST] = { .type = NLA_U32 }, +}; + +static int tbf_change(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + int err; + struct tbf_sched_data *q = qdisc_priv(sch); + struct nlattr *tb[TCA_TBF_MAX + 1]; + struct tc_tbf_qopt *qopt; + struct Qdisc *child = NULL; + struct psched_ratecfg rate; + struct psched_ratecfg peak; + u64 max_size; + s64 buffer, mtu; + u64 rate64 = 0, prate64 = 0; + + err = nla_parse_nested(tb, TCA_TBF_MAX, opt, tbf_policy, NULL); + if (err < 0) + return err; + + err = -EINVAL; + if (tb[TCA_TBF_PARMS] == NULL) + goto done; + + qopt = nla_data(tb[TCA_TBF_PARMS]); + if (qopt->rate.linklayer == TC_LINKLAYER_UNAWARE) + qdisc_put_rtab(qdisc_get_rtab(&qopt->rate, + tb[TCA_TBF_RTAB], + NULL)); + + if (qopt->peakrate.linklayer == TC_LINKLAYER_UNAWARE) + qdisc_put_rtab(qdisc_get_rtab(&qopt->peakrate, + tb[TCA_TBF_PTAB], + NULL)); + + buffer = min_t(u64, PSCHED_TICKS2NS(qopt->buffer), ~0U); + mtu = min_t(u64, PSCHED_TICKS2NS(qopt->mtu), ~0U); + + if (tb[TCA_TBF_RATE64]) + rate64 = nla_get_u64(tb[TCA_TBF_RATE64]); + psched_ratecfg_precompute(&rate, &qopt->rate, rate64); + + if (tb[TCA_TBF_BURST]) { + max_size = nla_get_u32(tb[TCA_TBF_BURST]); + buffer = psched_l2t_ns(&rate, max_size); + } else { + max_size = min_t(u64, psched_ns_t2l(&rate, buffer), ~0U); + } + + if (qopt->peakrate.rate) { + if (tb[TCA_TBF_PRATE64]) + prate64 = nla_get_u64(tb[TCA_TBF_PRATE64]); + psched_ratecfg_precompute(&peak, &qopt->peakrate, prate64); + if (peak.rate_bytes_ps <= rate.rate_bytes_ps) { + pr_warn_ratelimited("sch_tbf: peakrate %llu is lower than or equals to rate %llu !\n", + peak.rate_bytes_ps, rate.rate_bytes_ps); + err = -EINVAL; + goto done; + } + + if (tb[TCA_TBF_PBURST]) { + u32 pburst = nla_get_u32(tb[TCA_TBF_PBURST]); + max_size = min_t(u32, max_size, pburst); + mtu = psched_l2t_ns(&peak, pburst); + } else { + max_size = min_t(u64, max_size, psched_ns_t2l(&peak, mtu)); + } + } else { + memset(&peak, 0, sizeof(peak)); + } + + if (max_size < psched_mtu(qdisc_dev(sch))) + pr_warn_ratelimited("sch_tbf: burst %llu is lower than device %s mtu (%u) !\n", + max_size, qdisc_dev(sch)->name, + psched_mtu(qdisc_dev(sch))); + + if (!max_size) { + err = -EINVAL; + goto done; + } + + if (q->qdisc != &noop_qdisc) { + err = fifo_set_limit(q->qdisc, qopt->limit); + if (err) + goto done; + } else if (qopt->limit > 0) { + child = fifo_create_dflt(sch, &bfifo_qdisc_ops, qopt->limit, + extack); + if (IS_ERR(child)) { + err = PTR_ERR(child); + goto done; + } + + /* child is fifo, no need to check for noop_qdisc */ + qdisc_hash_add(child, true); + } + + sch_tree_lock(sch); + if (child) { + qdisc_tree_reduce_backlog(q->qdisc, q->qdisc->q.qlen, + q->qdisc->qstats.backlog); + qdisc_put(q->qdisc); + q->qdisc = child; + } + q->limit = qopt->limit; + if (tb[TCA_TBF_PBURST]) + q->mtu = mtu; + else + q->mtu = PSCHED_TICKS2NS(qopt->mtu); + q->max_size = max_size; + if (tb[TCA_TBF_BURST]) + q->buffer = buffer; + else + q->buffer = PSCHED_TICKS2NS(qopt->buffer); + q->tokens = q->buffer; + q->ptokens = q->mtu; + + memcpy(&q->rate, &rate, sizeof(struct psched_ratecfg)); + memcpy(&q->peak, &peak, sizeof(struct psched_ratecfg)); + + sch_tree_unlock(sch); + err = 0; +done: + return err; +} + +static int tbf_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + + qdisc_watchdog_init(&q->watchdog, sch); + q->qdisc = &noop_qdisc; + + if (!opt) + return -EINVAL; + + q->t_c = ktime_get_ns(); + + return tbf_change(sch, opt, extack); +} + +static void tbf_destroy(struct Qdisc *sch) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + + qdisc_watchdog_cancel(&q->watchdog); + qdisc_put(q->qdisc); +} + +static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + struct nlattr *nest; + struct tc_tbf_qopt opt; + + sch->qstats.backlog = q->qdisc->qstats.backlog; + nest = nla_nest_start(skb, TCA_OPTIONS); + if (nest == NULL) + goto nla_put_failure; + + opt.limit = q->limit; + psched_ratecfg_getrate(&opt.rate, &q->rate); + if (tbf_peak_present(q)) + psched_ratecfg_getrate(&opt.peakrate, &q->peak); + else + memset(&opt.peakrate, 0, sizeof(opt.peakrate)); + opt.mtu = PSCHED_NS2TICKS(q->mtu); + opt.buffer = PSCHED_NS2TICKS(q->buffer); + if (nla_put(skb, TCA_TBF_PARMS, sizeof(opt), &opt)) + goto nla_put_failure; + if (q->rate.rate_bytes_ps >= (1ULL << 32) && + nla_put_u64_64bit(skb, TCA_TBF_RATE64, q->rate.rate_bytes_ps, + TCA_TBF_PAD)) + goto nla_put_failure; + if (tbf_peak_present(q) && + q->peak.rate_bytes_ps >= (1ULL << 32) && + nla_put_u64_64bit(skb, TCA_TBF_PRATE64, q->peak.rate_bytes_ps, + TCA_TBF_PAD)) + goto nla_put_failure; + + return nla_nest_end(skb, nest); + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static int tbf_dump_class(struct Qdisc *sch, unsigned long cl, + struct sk_buff *skb, struct tcmsg *tcm) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + + tcm->tcm_handle |= TC_H_MIN(1); + tcm->tcm_info = q->qdisc->handle; + + return 0; +} + +static int tbf_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, + struct Qdisc **old, struct netlink_ext_ack *extack) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + + if (new == NULL) + new = &noop_qdisc; + + *old = qdisc_replace(sch, new, &q->qdisc); + return 0; +} + +static struct Qdisc *tbf_leaf(struct Qdisc *sch, unsigned long arg) +{ + struct tbf_sched_data *q = qdisc_priv(sch); + return q->qdisc; +} + +static unsigned long tbf_find(struct Qdisc *sch, u32 classid) +{ + return 1; +} + +static void tbf_walk(struct Qdisc *sch, struct qdisc_walker *walker) +{ + if (!walker->stop) { + if (walker->count >= walker->skip) + if (walker->fn(sch, 1, walker) < 0) { + walker->stop = 1; + return; + } + walker->count++; + } +} + +static const struct Qdisc_class_ops tbf_class_ops = { + .graft = tbf_graft, + .leaf = tbf_leaf, + .find = tbf_find, + .walk = tbf_walk, + .dump = tbf_dump_class, +}; + +static struct Qdisc_ops tbf_qdisc_ops __read_mostly = { + .next = NULL, + .cl_ops = &tbf_class_ops, + .id = "tbf", + .priv_size = sizeof(struct tbf_sched_data), + .enqueue = tbf_enqueue, + .dequeue = tbf_dequeue, + .peek = qdisc_peek_dequeued, + .init = tbf_init, + .reset = tbf_reset, + .destroy = tbf_destroy, + .change = tbf_change, + .dump = tbf_dump, + .owner = THIS_MODULE, +}; + +static int __init tbf_module_init(void) +{ + return register_qdisc(&tbf_qdisc_ops); +} + +static void __exit tbf_module_exit(void) +{ + unregister_qdisc(&tbf_qdisc_ops); +} +module_init(tbf_module_init) +module_exit(tbf_module_exit) +MODULE_LICENSE("GPL"); diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c new file mode 100644 index 000000000..163364124 --- /dev/null +++ b/net/sched/sch_teql.c @@ -0,0 +1,530 @@ +/* net/sched/sch_teql.c "True" (or "trivial") link equalizer. + * + * 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. + * + * Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/if_arp.h> +#include <linux/netdevice.h> +#include <linux/init.h> +#include <linux/skbuff.h> +#include <linux/moduleparam.h> +#include <net/dst.h> +#include <net/neighbour.h> +#include <net/pkt_sched.h> + +/* + How to setup it. + ---------------- + + After loading this module you will find a new device teqlN + and new qdisc with the same name. To join a slave to the equalizer + you should just set this qdisc on a device f.e. + + # tc qdisc add dev eth0 root teql0 + # tc qdisc add dev eth1 root teql0 + + That's all. Full PnP 8) + + Applicability. + -------------- + + 1. Slave devices MUST be active devices, i.e., they must raise the tbusy + signal and generate EOI events. If you want to equalize virtual devices + like tunnels, use a normal eql device. + 2. This device puts no limitations on physical slave characteristics + f.e. it will equalize 9600baud line and 100Mb ethernet perfectly :-) + Certainly, large difference in link speeds will make the resulting + eqalized link unusable, because of huge packet reordering. + I estimate an upper useful difference as ~10 times. + 3. If the slave requires address resolution, only protocols using + neighbour cache (IPv4/IPv6) will work over the equalized link. + Other protocols are still allowed to use the slave device directly, + which will not break load balancing, though native slave + traffic will have the highest priority. */ + +struct teql_master { + struct Qdisc_ops qops; + struct net_device *dev; + struct Qdisc *slaves; + struct list_head master_list; + unsigned long tx_bytes; + unsigned long tx_packets; + unsigned long tx_errors; + unsigned long tx_dropped; +}; + +struct teql_sched_data { + struct Qdisc *next; + struct teql_master *m; + struct sk_buff_head q; +}; + +#define NEXT_SLAVE(q) (((struct teql_sched_data *)qdisc_priv(q))->next) + +#define FMASK (IFF_BROADCAST | IFF_POINTOPOINT) + +/* "teql*" qdisc routines */ + +static int +teql_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) +{ + struct net_device *dev = qdisc_dev(sch); + struct teql_sched_data *q = qdisc_priv(sch); + + if (q->q.qlen < dev->tx_queue_len) { + __skb_queue_tail(&q->q, skb); + return NET_XMIT_SUCCESS; + } + + return qdisc_drop(skb, sch, to_free); +} + +static struct sk_buff * +teql_dequeue(struct Qdisc *sch) +{ + struct teql_sched_data *dat = qdisc_priv(sch); + struct netdev_queue *dat_queue; + struct sk_buff *skb; + struct Qdisc *q; + + skb = __skb_dequeue(&dat->q); + dat_queue = netdev_get_tx_queue(dat->m->dev, 0); + q = rcu_dereference_bh(dat_queue->qdisc); + + if (skb == NULL) { + struct net_device *m = qdisc_dev(q); + if (m) { + dat->m->slaves = sch; + netif_wake_queue(m); + } + } else { + qdisc_bstats_update(sch, skb); + } + sch->q.qlen = dat->q.qlen + q->q.qlen; + return skb; +} + +static struct sk_buff * +teql_peek(struct Qdisc *sch) +{ + /* teql is meant to be used as root qdisc */ + return NULL; +} + +static void +teql_reset(struct Qdisc *sch) +{ + struct teql_sched_data *dat = qdisc_priv(sch); + + skb_queue_purge(&dat->q); + sch->q.qlen = 0; +} + +static void +teql_destroy(struct Qdisc *sch) +{ + struct Qdisc *q, *prev; + struct teql_sched_data *dat = qdisc_priv(sch); + struct teql_master *master = dat->m; + + if (!master) + return; + + prev = master->slaves; + if (prev) { + do { + q = NEXT_SLAVE(prev); + if (q == sch) { + NEXT_SLAVE(prev) = NEXT_SLAVE(q); + if (q == master->slaves) { + master->slaves = NEXT_SLAVE(q); + if (q == master->slaves) { + struct netdev_queue *txq; + spinlock_t *root_lock; + + txq = netdev_get_tx_queue(master->dev, 0); + master->slaves = NULL; + + root_lock = qdisc_root_sleeping_lock(rtnl_dereference(txq->qdisc)); + spin_lock_bh(root_lock); + qdisc_reset(rtnl_dereference(txq->qdisc)); + spin_unlock_bh(root_lock); + } + } + skb_queue_purge(&dat->q); + break; + } + + } while ((prev = q) != master->slaves); + } +} + +static int teql_qdisc_init(struct Qdisc *sch, struct nlattr *opt, + struct netlink_ext_ack *extack) +{ + struct net_device *dev = qdisc_dev(sch); + struct teql_master *m = (struct teql_master *)sch->ops; + struct teql_sched_data *q = qdisc_priv(sch); + + if (dev->hard_header_len > m->dev->hard_header_len) + return -EINVAL; + + if (m->dev == dev) + return -ELOOP; + + q->m = m; + + skb_queue_head_init(&q->q); + + if (m->slaves) { + if (m->dev->flags & IFF_UP) { + if ((m->dev->flags & IFF_POINTOPOINT && + !(dev->flags & IFF_POINTOPOINT)) || + (m->dev->flags & IFF_BROADCAST && + !(dev->flags & IFF_BROADCAST)) || + (m->dev->flags & IFF_MULTICAST && + !(dev->flags & IFF_MULTICAST)) || + dev->mtu < m->dev->mtu) + return -EINVAL; + } else { + if (!(dev->flags&IFF_POINTOPOINT)) + m->dev->flags &= ~IFF_POINTOPOINT; + if (!(dev->flags&IFF_BROADCAST)) + m->dev->flags &= ~IFF_BROADCAST; + if (!(dev->flags&IFF_MULTICAST)) + m->dev->flags &= ~IFF_MULTICAST; + if (dev->mtu < m->dev->mtu) + m->dev->mtu = dev->mtu; + } + q->next = NEXT_SLAVE(m->slaves); + NEXT_SLAVE(m->slaves) = sch; + } else { + q->next = sch; + m->slaves = sch; + m->dev->mtu = dev->mtu; + m->dev->flags = (m->dev->flags&~FMASK)|(dev->flags&FMASK); + } + return 0; +} + + +static int +__teql_resolve(struct sk_buff *skb, struct sk_buff *skb_res, + struct net_device *dev, struct netdev_queue *txq, + struct dst_entry *dst) +{ + struct neighbour *n; + int err = 0; + + n = dst_neigh_lookup_skb(dst, skb); + if (!n) + return -ENOENT; + + if (dst->dev != dev) { + struct neighbour *mn; + + mn = __neigh_lookup_errno(n->tbl, n->primary_key, dev); + neigh_release(n); + if (IS_ERR(mn)) + return PTR_ERR(mn); + n = mn; + } + + if (neigh_event_send(n, skb_res) == 0) { + int err; + char haddr[MAX_ADDR_LEN]; + + neigh_ha_snapshot(haddr, n, dev); + err = dev_hard_header(skb, dev, ntohs(skb_protocol(skb, false)), + haddr, NULL, skb->len); + + if (err < 0) + err = -EINVAL; + } else { + err = (skb_res == NULL) ? -EAGAIN : 1; + } + neigh_release(n); + return err; +} + +static inline int teql_resolve(struct sk_buff *skb, + struct sk_buff *skb_res, + struct net_device *dev, + struct netdev_queue *txq) +{ + struct dst_entry *dst = skb_dst(skb); + int res; + + if (rcu_access_pointer(txq->qdisc) == &noop_qdisc) + return -ENODEV; + + if (!dev->header_ops || !dst) + return 0; + + rcu_read_lock(); + res = __teql_resolve(skb, skb_res, dev, txq, dst); + rcu_read_unlock(); + + return res; +} + +static netdev_tx_t teql_master_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct teql_master *master = netdev_priv(dev); + struct Qdisc *start, *q; + int busy; + int nores; + int subq = skb_get_queue_mapping(skb); + struct sk_buff *skb_res = NULL; + + start = master->slaves; + +restart: + nores = 0; + busy = 0; + + q = start; + if (!q) + goto drop; + + do { + struct net_device *slave = qdisc_dev(q); + struct netdev_queue *slave_txq = netdev_get_tx_queue(slave, 0); + + if (slave_txq->qdisc_sleeping != q) + continue; + if (netif_xmit_stopped(netdev_get_tx_queue(slave, subq)) || + !netif_running(slave)) { + busy = 1; + continue; + } + + switch (teql_resolve(skb, skb_res, slave, slave_txq)) { + case 0: + if (__netif_tx_trylock(slave_txq)) { + unsigned int length = qdisc_pkt_len(skb); + + if (!netif_xmit_frozen_or_stopped(slave_txq) && + netdev_start_xmit(skb, slave, slave_txq, false) == + NETDEV_TX_OK) { + __netif_tx_unlock(slave_txq); + master->slaves = NEXT_SLAVE(q); + netif_wake_queue(dev); + master->tx_packets++; + master->tx_bytes += length; + return NETDEV_TX_OK; + } + __netif_tx_unlock(slave_txq); + } + if (netif_xmit_stopped(netdev_get_tx_queue(dev, 0))) + busy = 1; + break; + case 1: + master->slaves = NEXT_SLAVE(q); + return NETDEV_TX_OK; + default: + nores = 1; + break; + } + __skb_pull(skb, skb_network_offset(skb)); + } while ((q = NEXT_SLAVE(q)) != start); + + if (nores && skb_res == NULL) { + skb_res = skb; + goto restart; + } + + if (busy) { + netif_stop_queue(dev); + return NETDEV_TX_BUSY; + } + master->tx_errors++; + +drop: + master->tx_dropped++; + dev_kfree_skb(skb); + return NETDEV_TX_OK; +} + +static int teql_master_open(struct net_device *dev) +{ + struct Qdisc *q; + struct teql_master *m = netdev_priv(dev); + int mtu = 0xFFFE; + unsigned int flags = IFF_NOARP | IFF_MULTICAST; + + if (m->slaves == NULL) + return -EUNATCH; + + flags = FMASK; + + q = m->slaves; + do { + struct net_device *slave = qdisc_dev(q); + + if (slave == NULL) + return -EUNATCH; + + if (slave->mtu < mtu) + mtu = slave->mtu; + if (slave->hard_header_len > LL_MAX_HEADER) + return -EINVAL; + + /* If all the slaves are BROADCAST, master is BROADCAST + If all the slaves are PtP, master is PtP + Otherwise, master is NBMA. + */ + if (!(slave->flags&IFF_POINTOPOINT)) + flags &= ~IFF_POINTOPOINT; + if (!(slave->flags&IFF_BROADCAST)) + flags &= ~IFF_BROADCAST; + if (!(slave->flags&IFF_MULTICAST)) + flags &= ~IFF_MULTICAST; + } while ((q = NEXT_SLAVE(q)) != m->slaves); + + m->dev->mtu = mtu; + m->dev->flags = (m->dev->flags&~FMASK) | flags; + netif_start_queue(m->dev); + return 0; +} + +static int teql_master_close(struct net_device *dev) +{ + netif_stop_queue(dev); + return 0; +} + +static void teql_master_stats64(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct teql_master *m = netdev_priv(dev); + + stats->tx_packets = m->tx_packets; + stats->tx_bytes = m->tx_bytes; + stats->tx_errors = m->tx_errors; + stats->tx_dropped = m->tx_dropped; +} + +static int teql_master_mtu(struct net_device *dev, int new_mtu) +{ + struct teql_master *m = netdev_priv(dev); + struct Qdisc *q; + + q = m->slaves; + if (q) { + do { + if (new_mtu > qdisc_dev(q)->mtu) + return -EINVAL; + } while ((q = NEXT_SLAVE(q)) != m->slaves); + } + + dev->mtu = new_mtu; + return 0; +} + +static const struct net_device_ops teql_netdev_ops = { + .ndo_open = teql_master_open, + .ndo_stop = teql_master_close, + .ndo_start_xmit = teql_master_xmit, + .ndo_get_stats64 = teql_master_stats64, + .ndo_change_mtu = teql_master_mtu, +}; + +static __init void teql_master_setup(struct net_device *dev) +{ + struct teql_master *master = netdev_priv(dev); + struct Qdisc_ops *ops = &master->qops; + + master->dev = dev; + ops->priv_size = sizeof(struct teql_sched_data); + + ops->enqueue = teql_enqueue; + ops->dequeue = teql_dequeue; + ops->peek = teql_peek; + ops->init = teql_qdisc_init; + ops->reset = teql_reset; + ops->destroy = teql_destroy; + ops->owner = THIS_MODULE; + + dev->netdev_ops = &teql_netdev_ops; + dev->type = ARPHRD_VOID; + dev->mtu = 1500; + dev->min_mtu = 68; + dev->max_mtu = 65535; + dev->tx_queue_len = 100; + dev->flags = IFF_NOARP; + dev->hard_header_len = LL_MAX_HEADER; + netif_keep_dst(dev); +} + +static LIST_HEAD(master_dev_list); +static int max_equalizers = 1; +module_param(max_equalizers, int, 0); +MODULE_PARM_DESC(max_equalizers, "Max number of link equalizers"); + +static int __init teql_init(void) +{ + int i; + int err = -ENODEV; + + for (i = 0; i < max_equalizers; i++) { + struct net_device *dev; + struct teql_master *master; + + dev = alloc_netdev(sizeof(struct teql_master), "teql%d", + NET_NAME_UNKNOWN, teql_master_setup); + if (!dev) { + err = -ENOMEM; + break; + } + + if ((err = register_netdev(dev))) { + free_netdev(dev); + break; + } + + master = netdev_priv(dev); + + strlcpy(master->qops.id, dev->name, IFNAMSIZ); + err = register_qdisc(&master->qops); + + if (err) { + unregister_netdev(dev); + free_netdev(dev); + break; + } + + list_add_tail(&master->master_list, &master_dev_list); + } + return i ? 0 : err; +} + +static void __exit teql_exit(void) +{ + struct teql_master *master, *nxt; + + list_for_each_entry_safe(master, nxt, &master_dev_list, master_list) { + + list_del(&master->master_list); + + unregister_qdisc(&master->qops); + unregister_netdev(master->dev); + free_netdev(master->dev); + } +} + +module_init(teql_init); +module_exit(teql_exit); + +MODULE_LICENSE("GPL"); |