CN102857445B - Low-expenditure distributing structure and distributing method of network-on-chip router - Google Patents
Low-expenditure distributing structure and distributing method of network-on-chip router Download PDFInfo
- Publication number
- CN102857445B CN102857445B CN201210332337.3A CN201210332337A CN102857445B CN 102857445 B CN102857445 B CN 102857445B CN 201210332337 A CN201210332337 A CN 201210332337A CN 102857445 B CN102857445 B CN 102857445B
- Authority
- CN
- China
- Prior art keywords
- virtual channel
- request
- distribution
- module
- successfully
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Abstract
The invention discloses a low-expenditure distributing structure and distributing method of a network-on-chip router, which aim to solve the problems of high proportion of resources occupied by virtual channel distributing units, repeated work of the distribution of virtual channels and the distribution of switches, and long time delay in distribution in the distribution of virtual channels and the distribution of switches in the traditional virtual channel router. The low-expenditure distributing structure of a network-on-chip router comprises a first distributing structure and a second distributing structure, wherein the first distributing structure is distributed in output ports; each distributing unit of output ports comprises an invalid new request removing module, a PV: 1 arbitrator module and a simple virtual channel distributing module, and the first distributing structure completes the first step of arbitration and virtual channel distribution of the switch distribution. The second distributing structure is positioned in all input ports, each distributing unit of input ports is a V: 1 arbitrator module; the second distribution structure completes the second step of arbitration of the switch distribution; and the second distribution structure and the first distribution structure together complete the operation of virtual channel distribution and switch distribution of the virtual channel router. The low-expenditure distributing structure and the distributing method of a network-on-chip router have the advantages of short time delay and small size.
Description
Technical field
The invention belongs to communication technical field, be specifically related to a kind of low expense distribution structure and distribution method of the network-on-chip router for the distribution of router virtual channel and switch distribution.
Background technology
Along with CMOS technique is stepped into nanoscale, can integrated billions of transistors on one single chip, SOC (system on a chip) (SoC) design becomes increasingly complex, and the integrated IP kernel of on-chip system chip is also more and more, and the communication issue of how handling well between multinuclear is paid close attention to day by day.Traditional bus structures are seriously restricting performance boost, because bus structures exist following two aspect bottlenecks: (1) limit bandwidth.Although bus can effectively connect multiple IP kernels, bus address resource can not be along with the increase of IP kernel infinite expanding, and bus cannot support more than one pair of IP kernel to communicate by letter simultaneously, in the time of a bus of multiple IP kernels competition, can cause serious obstruction.(2) clock problem.Along with integration characteristic size enters nanoscale, the physical effect problem that nanometer technology brings will make on sheet global synchronization more and more difficult, and clock will produce deflection and become uncontrollable, and chip cannot normally be worked.
In recent years, in order to overcome the problems referred to above, use for reference the thought of computer network, researcher is incorporated into the concept of network in chip, and network-on-chip (NoC) structure of proposition has effectively solved bus-structured problems.Network-on-chip mainly comprises router, the interconnecting link between router and network interface, and IP kernel is connected with router by network interface.The transmission of data in network-on-chip must forward by router, and therefore router is the key that affects network-on-chip performance.
Tradition virtual channel router comprises P input and output unit, a route computing unit, virtual channel allocation units, switch allocation units and a cross bar switch.Tradition virtual channel router adopts 4 level production line designs, is respectively that route is calculated, virtual channel distributes, switch distributes and switch transmission.
First route computing unit is that data calculate the output port that will go to, and then virtual channel allocation units could be the virtual channel that this output port of this data allocations can be used.After virtual channel is allocated successfully, carry out again switch distribution, for this data configuration cross bar switch, link corresponding input port and output port.Finally carry out switch transmission according to the passage of configuration successful, leave current router.Virtual channel distributes and distributes with switch is requisite two steps of traditional virtual channel router transmission data, and virtual channel distributes with switch and distributes and respectively account for a level production line, therefore they distribute needed time delay to account for microplate by the half of needed total time of router, time delay proportion is larger, and router time delay is one of important parameter of weighing router performance.
The virtual channel of tradition virtual channel router distributes and can multiple output virtual channels of same output port be distributed to different input virtual channel requests simultaneously, and two steps arbitration principles are followed in switch distribution, it is to arbitrate one from multiple output virtual channels that become the distribution of work of same output port that its second stage is distributed, this has caused virtual channel to share out the work for switch distributes, and has repeatability.
The virtual channel of tradition virtual channel router distributes always total P*V P*V:1 moderator of second stage, and the size of moderator and number all can be along with the change of port number or virtual channel number increases greatly and sharply, and therefore virtual channel distributes shared router resource proportion larger.
Summary of the invention
The object of the invention is to for above-mentioned the deficiencies in the prior art, a kind of low expense distribution structure and distribution method of network-on-chip router are proposed, this structure merges virtual channel allocation units and switch allocation units, to solve resource that traditional virtual channel router virtual channel allocation units take than great, virtual channel distributes with switch assignment repetition and virtual channel and distributes the problem that the distribution time delay of distributing with switch is large.
For achieving the above object, the low expense distribution structure of network-on-chip router of the present invention is: be made up of two step distribution structures, wherein first step distribution structure is distributed in output port, comprise P output port allocation units, second step distribution structure is distributed in input port, comprise P input port allocation units, described P input of output port allocation units and the output of route computing unit are connected, output is connected with the input of P input port allocation units, and the output of described P input port allocation units is connected with cross bar switch;
Described each output port allocation units include an invalid new request module of filtering, a PV:1 moderator module and a simple virtual channel distribution module;
Described each input port allocation units are a V:1 moderator module;
The invalid new request module of described filtering receives the input virtual channel request of route computing unit, and invalid request in these requests of filtering, and remaining effective request is sent into PV:1 moderator module;
The effective request of described PV:1 moderator module after to the invalid new request module filtering of filtering carried out switch and distributed first step arbitration, and competition is successfully asked to send into simple virtual channel distribution module;
Virtual channel allocation process is carried out in the competition that described simple virtual channel distribution module sends according to the PV:1 moderator module successfully character of request, result is inputed to V:1 moderator module and complete the second step arbitration that switch distributes.
The low expense distribution method of network-on-chip router of the present invention comprises the following steps:
1) first stage distributes, and completes first step arbitration and virtual channel distribution that switch distributes:
1a) the invalid new request module of filtering receives the virtual channel request of being inputted by route computing unit, and call the virtual channel assignment information of simple virtual channel distribution module record, according to the type of the virtual channel assignment information judgement input virtual channel request of record, unsuccessful request marks of carrying out virtual channel distribution is " newly request ", and the request marks of successfully carrying out virtual channel distribution is " successfully asking ";
Whether available free output virtual channel also of the output port that 1b) the invalid new request module of filtering is asked according to output virtual channel state information judgement " newly request ", if there is no idle output virtual channel, represent " newly request " to be invalid new request, should " newly ask " filtering, if available free output virtual channel, is left intact to this " newly request ", " the newly request " of remaining not filtering and " successfully request " are sent to PV:1 moderator module;
1c) PV:1 moderator module receives effective request that the invalid new request module of filtering is sent into, and these requests are arbitrated, and completes switch and distributes the first step, and the competition successfully input virtual channel numbers of request is sent into simple virtual channel distribution module as arbitration result;
1d) simple virtual channel distribution module receives the arbitration result that PV:1 moderator module is sent here, judge that it is " newly request " or " successfully request " that virtual channel request is successfully inputted in this competition, if " newly request ", by fixing order, first idle output virtual channel is distributed to this " newly request ", record virtual channel assignment information, the output virtual channel state information that renewal is assigned with is down state, if " successfully request " skips the operation that virtual channel distributes; The arbitration result of processing through judgement is sent to V:1 moderator module;
2) carry out second stage distribution, complete the second step arbitration that switch distributes: the arbitration result that the simple virtual channel distribution module of P output port of V:1 moderator module reception is sent here, these arbitration result are carried out to the second step arbitration of switch distribution, finally complete global switch and distribute.
The present invention compared with prior art has the following advantages:
1, the virtual channel distributor of existing traditional virtual channel router and switch distributor have P*V P*V:1 moderator, V V:1 moderator of (P+1) * and P P:1 moderator, and distribution structure of the present invention has P V:1 moderator and P P*V:1 moderator, filtering invalidation request module in the present invention, distinguish new request fewer with old request module and the new shared resource of virtual channel allocation units, therefore moderator of the present invention is number with size all than little many of traditional virtual channel router, has reduced area.
2, the virtual channel of existing traditional virtual channel router distributes and switch assignment existence repeatability, and simple virtual channel distribution module only distributes the single request of first step arbitration triumph to carry out virtual channel distribution to switch in the present invention, therefore at most only dispense an output virtual channel, the repeatability of having avoided virtual channel to share out the work at every turn.
3, existing traditional virtual channel router adopts pipelining, wherein virtual channel distribution distributes and respectively accounts for a level production line with switch, virtual channel distribution simultaneously distributes and all follows two_phase assignment principle with switch, and the present invention adopts simple simple virtual channel distribution module to replace traditional virtual channel allocation units, and this simple virtual channel distribution module is dissolved in the first stage of switch distribution, thereby distributing virtual channel to distribute with switch virtual channel distributes and merges into a level production line, reduced distribution time delay and improved router delay performance.
Brief description of the drawings
Fig. 1 is virtual channel router topology schematic diagram of the present invention;
Fig. 2 is the low expense distribution structure schematic diagram of network-on-chip router of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described further.
Referring to Fig. 1, virtual channel router of the present invention comprises P input unit 101, a route computing unit 102, low expense allocation units 103 and a cross bar switch 104, be with traditional virtual channel router difference, traditional virtual channel router is by P input unit, a route computing unit, virtual channel allocation units, switch allocation units and a cross bar switch composition, wherein virtual channel distribution is two modules that separate with switch distribution, virtual channel allocation units and switch allocation units have formed the distribution structure of traditional virtual channel router, and the present invention directly distributes virtual channel with switch and distributes and merge into the i.e. low expense allocation units 103 of a module.
With reference to Fig. 2, the low expense distribution structure of network-on-chip router of the present invention, virtual channel is distributed with switch and distributes and merge into one-level pile line operation, formed by two step distribution structures, wherein first step distribution structure is distributed in output port, comprise P output port allocation units 201, second step distribution structure is distributed in input port, comprise P input port allocation units 205, the input of described P output port allocation units 201 is connected with the output of route computing unit 102, output is connected with the input of P input port allocation units 205, the output of described P input port allocation units 205 is connected with cross bar switch 104,
Taking one of them output port allocation units 201 as example, it comprises an invalid new request module 202 of filtering, a PV:1 moderator module 203 and a simple virtual channel distribution module 204, wherein the invalid new request module 202 of filtering is connected with PV:1 moderator module 203 by effective solicited message, and PV:1 moderator module 203 is connected with simple virtual channel distribution module 204 by the successful solicited message of competition again;
The invalid new request module 202 of described filtering receives the virtual channel request that route computing unit 102 is inputted, and invalid request in these requests of filtering, and remaining effective request is sent into PV:1 moderator module 203.
Described PV:1 moderator module 203 is effectively asked to carry out switch to these and is distributed first step arbitration, and successful competition request is sent into simple virtual channel distribution module 204.
The character that described simple virtual channel distribution module 204 is successfully asked according to this competition is carried out virtual channel allocation process, result is sent to the V:1 moderator module 206 of input port allocation units 205;
Described each input port allocation units 205 are a V:1 moderator module 206, taking one of them V:1 moderator module 206 as example, the character of P the successful request of competition that simply virtual channel distribution module 204 sends according to PV:1 moderator module 203 is carried out virtual channel allocation process, result is inputed to P V:1 moderator module 206 and complete the second step arbitration that switch distributes;
The first step arbitration that described switch distributes refers to all output ports in the same time by a PV:1 moderator separately, from the input virtual channel request of all requests oneself, selects one;
The second step arbitration that described switch distributes refers to all input ports in the same time by a V:1 moderator separately, from the input virtual channel request of oneself, selects one.
The allocation step that the present invention realizes the low expense distribution structure of network-on-chip router has:
1) first stage distributes, and taking output port allocation units 201 as example, the process that the first step is arbitrated and virtual channel distributes that completes switch distribution is as follows:
1a) the invalid new request module 202 of filtering receives the virtual channel request of being inputted by route computing unit 101, and call the virtual channel assignment information that simple virtual channel distribution module 204 records, according to the type of the virtual channel assignment information judgement input virtual channel request of record, unsuccessful request marks of carrying out virtual channel distribution is " newly request ", and the request marks of successfully carrying out virtual channel distribution is " successfully asking ";
Whether available free output virtual channel also of the output port that 1b) the invalid new request module 202 of filtering is asked according to output virtual channel state information judgement " newly request ", if there is no idle output virtual channel, represent " newly request " to be invalid new request, should " newly ask " filtering, if available free output virtual channel, is left intact to this " newly request ", " the newly request " of remaining not filtering and " successfully request " are sent to PV:1 moderator module 203;
1c) PV:1 moderator module 203 receives effective request that the invalid new request module 202 of filtering is sent into, these requests are arbitrated, complete switch and distribute the first step, the input virtual channel numbers that competition is successfully asked is sent into simple virtual channel distribution module 204 as arbitration result;
1d) simple virtual channel distribution module 204 receives the arbitration result that PV:1 moderator module 203 is sent here, judge that it is " newly request " or " successfully request " that virtual channel request is successfully inputted in this competition, if " newly request ", by fixing order, first idle output virtual channel is distributed to this " newly request ", record virtual channel assignment information, the output virtual channel state information that renewal is assigned with is down state, if " successfully request " skips the operation that virtual channel distributes; The arbitration result of processing through judgement is sent to V:1 moderator module 206;
1e) remaining all output port allocation units 201 also perform step the 1a in 1 in the same time in order), 1b), 1c) and 1d) step by step, jointly complete the distribution of first stage;
Step 2, carries out second stage distribution, taking input port allocation units 205 as example, completes the second step arbitration that switch distributes:
2a) V:1 moderator module 206 receives P the input virtual channel numbers that simple virtual channel distribution module 203 is sent here of first step distribution structure, these input virtual channel numbers are carried out to second stage switch and distribute arbitration, the arbitration result obtaining is as the read signal of buffer memory, relatively read signal and P input virtual channel numbers, if arbitration result is identical with one of them, just represent that input port 1 is finally connected with the output port of identical result representative, has completed global switch distribution;
2b) remaining all input port allocation units 205 also perform step in 2 2a step by step in the same time) operation, jointly complete second stage and distribute.
Claims (2)
1. the low expense distribution structure of a network-on-chip router, it is characterized in that: formed by two distribution structures, wherein the first distribution structure is distributed in output port, comprise P output port allocation units (201), the second distribution structure is distributed in input port, comprise P input port allocation units (205), the input of described P output port allocation units (201) is connected with the output of route computing unit (102), output is connected with the input of P input port allocation units (205), the output of described P input port allocation units (205) is connected with cross bar switch (104),
Described each output port allocation units (201) include an invalid new request module of filtering (202), a PV:1 moderator module (203) and a simple virtual channel distribution module (204);
Described each input port allocation units (205) are a V:1 moderator module (206);
The invalid new request module of described filtering (202) receives the input virtual channel request of route computing unit (102), and invalid request in these requests of filtering, remaining effective request is sent into PV:1 moderator module (203);
The effective request of described PV:1 moderator module (203) after to the invalid new request module of filtering (202) filtering carried out switch and distributed first step arbitration, and competition is successfully asked to send into simple virtual channel distribution module (204);
Virtual channel allocation process is carried out in the competition that described simple virtual channel distribution module (204) sends according to PV:1 moderator module (203) the successfully character of request, result is inputed to V:1 moderator module (206) and complete the second step arbitration that switch distributes; Concrete, simple virtual channel distribution module (204) receives the arbitration result that PV:1 moderator module (203) is sent here, judge that it is " newly request " or " successfully request " that virtual channel request is successfully inputted in this competition, if " newly request ", by fixing order, first idle output virtual channel is distributed to this " newly request ", record virtual channel assignment information, the output virtual channel state information that renewal is assigned with is down state, if " successfully request " skips the operation that virtual channel distributes; The arbitration result of processing through judgement is sent to V:1 moderator module (206); Wherein, the invalid new request module of filtering (202) is according to the type of the virtual channel assignment information judgement input virtual channel request of record, unsuccessful request marks of carrying out virtual channel distribution is " newly request ", and the request marks of successfully carrying out virtual channel distribution is " successfully asking ".
2. a distribution method for the low expense distribution structure of network-on-chip router as claimed in claim 1, is characterized in that comprising the following steps:
1) first stage distributes, and completes first step arbitration and virtual channel distribution that switch distributes:
1a) the invalid new request module of filtering (202) receives the virtual channel request by route computing unit (102) input, and call the virtual channel assignment information of simple virtual channel distribution module (204) record, according to the type of the virtual channel assignment information judgement input virtual channel request of record, unsuccessful request marks of carrying out virtual channel distribution is " newly request ", and the request marks of successfully carrying out virtual channel distribution is " successfully asking ";
Whether available free output virtual channel also of the output port that 1b) the invalid new request module of filtering (202) is asked according to output virtual channel state information judgement " newly request ", if there is no idle output virtual channel, represent " newly request " to be invalid new request, should " newly ask " filtering, if available free output virtual channel, is left intact to this " newly request ", " the newly request " of remaining not filtering and " successfully request " are sent to PV:1 moderator module (203);
1c) PV:1 moderator module (203) receives effective request that the invalid new request module of filtering (202) is sent into, these requests are arbitrated, complete switch and distribute the first step, the competition successfully input virtual channel numbers of request is sent into simple virtual channel distribution module (204) as arbitration result;
1d) simple virtual channel distribution module (204) receives the arbitration result that PV:1 moderator module (203) is sent here, judge that it is " newly request " or " successfully request " that virtual channel request is successfully inputted in this competition, if " newly request ", by fixing order, first idle output virtual channel is distributed to this " newly request ", record virtual channel assignment information, the output virtual channel state information that renewal is assigned with is down state, if " successfully request " skips the operation that virtual channel distributes; The arbitration result of processing through judgement is sent to V:1 moderator module (206);
2) carry out second stage distribution, complete the second step arbitration that switch distributes: the arbitration result that the simple virtual channel distribution module (203) of P output port of V:1 moderator module (206) reception is sent here, these arbitration result are carried out to the second step arbitration of switch distribution, finally complete global switch and distribute.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210332337.3A CN102857445B (en) | 2012-09-10 | 2012-09-10 | Low-expenditure distributing structure and distributing method of network-on-chip router |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210332337.3A CN102857445B (en) | 2012-09-10 | 2012-09-10 | Low-expenditure distributing structure and distributing method of network-on-chip router |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102857445A CN102857445A (en) | 2013-01-02 |
| CN102857445B true CN102857445B (en) | 2014-12-03 |
Family
ID=47403647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210332337.3A Expired - Fee Related CN102857445B (en) | 2012-09-10 | 2012-09-10 | Low-expenditure distributing structure and distributing method of network-on-chip router |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102857445B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550159B (en) * | 2015-12-16 | 2018-08-31 | 深圳先进技术研究院 | The power distribution method of multi-core processor network-on-chip |
| CN105721355A (en) * | 2016-01-29 | 2016-06-29 | 浪潮(北京)电子信息产业有限公司 | Method for transmitting message through network-on-chip route and network-on-chip route |
| CN106453072A (en) * | 2016-06-22 | 2017-02-22 | 中国科学院计算技术研究所 | Greedy distribution method and device of on-chip network router channel resources and router |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101232456A (en) * | 2008-01-25 | 2008-07-30 | 浙江大学 | Distributed type testing on-chip network router |
| CN102629913A (en) * | 2012-04-11 | 2012-08-08 | 浙江大学 | Router device suitable for globally asynchronous locally synchronous on-chip network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101495811B1 (en) * | 2008-06-09 | 2015-02-25 | 삼성전자주식회사 | Apparatus and method for high speed packet routing system |
-
2012
- 2012-09-10 CN CN201210332337.3A patent/CN102857445B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101232456A (en) * | 2008-01-25 | 2008-07-30 | 浙江大学 | Distributed type testing on-chip network router |
| CN102629913A (en) * | 2012-04-11 | 2012-08-08 | 浙江大学 | Router device suitable for globally asynchronous locally synchronous on-chip network |
Non-Patent Citations (2)
| Title |
|---|
| Li-Shiuan Peh,William J. Dally.A Delay Model and Speculative Architecture for Pipelined Routers.《IEEE-High-Performance Computer Architecture, 2001. HPCA》.2001,全文. * |
| Li-ShiuanPeh，WilliamJ.Dally.ADelayModelandSpeculativeArchitectureforPipelinedRouters.《IEEE-High-PerformanceComputerArchitecture 2001. HPCA》.2001 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102857445A (en) | 2013-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103248526B (en) | Realize signal equipment, the method for out-of-band supervision control management | |
| CN100499556C (en) | High-speed asynchronous interlinkage communication network of heterogeneous multi-nucleus processor | |
| CN101808032B (en) | Static XY routing algorithm-oriented two-dimensional grid NoC router optimization design method | |
| KR100812225B1 (en) | Crossbar switch architecture for multi-processor SoC platform | |
| US8325761B2 (en) | System and method for establishing sufficient virtual channel performance in a parallel computing network | |
| CN103729331B (en) | Network-on-chip communication deadlock avoiding method, router and communication network | |
| CN102685017A (en) | On-chip network router based on field programmable gate array (FPGA) | |
| CN204808315U (en) | System based on inter -plate communication is realized to SPI serial peripheral hardware interface protocol | |
| CN104901899B (en) | The adaptive routing method of two-dimentional Survey on network-on-chip topology | |
| CN102811152B (en) | Method for realizing real-time transaction and data exchange of multiple main bus network communication | |
| CN106155959A (en) | Data transmission method and data transmission system | |
| CN107766267A (en) | A kind of referee method and system of I2C buses | |
| CN105022717A (en) | Network on chip resource arbitration method and arbitration unit of additional request number priority | |
| CN103840993A (en) | Dual-redundancy CAN bus data transmission method | |
| CN102857445B (en) | Low-expenditure distributing structure and distributing method of network-on-chip router | |
| CN203590251U (en) | FlexRay control system based on serial RapidIO bus | |
| CN102035688A (en) | Design method for rapidly controlling network link access | |
| CN106506625A (en) | Four control servers and four control server communications methods | |
| CN101303677B (en) | Method and system for controlling accessing direct memory as well as controller | |
| CN105359122B (en) | enhanced data transmission in multi-CPU system | |
| CN103312614B (en) | A kind of multicast message processing method, line card and communication equipment | |
| CN105530157B (en) | The multiple virtual links of the AFDX network switch share the framework and method of credit | |
| CN103577469B (en) | Database connection multiplexing method and apparatus | |
| CN102299857A (en) | Multicast message processing method, line card and communication equipment | |
| US20160085706A1 (en) | Methods And Systems For Controlling Ordered Write Transactions To Multiple Devices Using Switch Point Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141203 Termination date: 20150910 |
|
| EXPY | Termination of patent right or utility model |