WO2010059150A1 - Interconnects using self-timed time-division multiplexed bus - Google Patents
Interconnects using self-timed time-division multiplexed bus Download PDFInfo
- Publication number
- WO2010059150A1 WO2010059150A1 PCT/US2008/083974 US2008083974W WO2010059150A1 WO 2010059150 A1 WO2010059150 A1 WO 2010059150A1 US 2008083974 W US2008083974 W US 2008083974W WO 2010059150 A1 WO2010059150 A1 WO 2010059150A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- clock
- unit
- transportation
- transportation unit
- Prior art date
Links
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000004891 communication Methods 0.000 claims abstract description 20
- 230000000977 initiatory effect Effects 0.000 claims abstract description 3
- 230000002457 bidirectional effect Effects 0.000 claims description 10
- 238000013481 data capture Methods 0.000 claims description 6
- 230000004048 modification Effects 0.000 claims description 6
- 238000012986 modification Methods 0.000 claims description 6
- 230000001902 propagating effect Effects 0.000 claims 2
- 238000013461 design Methods 0.000 description 14
- 230000003111 delayed effect Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 230000000630 rising effect Effects 0.000 description 8
- 101150066242 outO gene Proteins 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000001360 synchronised effect Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 229930091051 Arenine Natural products 0.000 description 1
- 241000723353 Chrysanthemum Species 0.000 description 1
- 235000005633 Chrysanthemum balsamita Nutrition 0.000 description 1
- 101100120298 Rattus norvegicus Flot1 gene Proteins 0.000 description 1
- 101100412403 Rattus norvegicus Reg3b gene Proteins 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 239000000872 buffer Substances 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000135 prohibitive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000004353 relayed correlation spectroscopy Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4204—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
- G06F13/4221—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus being an input/output bus, e.g. ISA bus, EISA bus, PCI bus, SCSI bus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L7/00—Arrangements for synchronising receiver with transmitter
- H04L7/0008—Synchronisation information channels, e.g. clock distribution lines
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Definitions
- This invention relates to the field of integrated circuits. More particularly, this invention relates to interconnection designs for integrated circuits.
- P2P point-to-point
- a point- to-point interconnection design provides a dedicated wire between every two points of the integrated circuit that need to be connected.
- point-to-point connections tend to have a relatively high number of interconnections in comparison to the number of points being connected.
- the longer wires not only slow down the interconnection but also consume more power, which results from two related effects of the longer wires, which are the higher wire load and the additional number of intermediate buffers that might be required to meet timing requirements.
- the increasing ratio of wire-delay versus gate- delay in deep sub-micron processes is a further detrimental factor. As the transistor feature size continues to shrink, the wire-shrink is not scaling well with it.
- Fig. IA is a unidirectional daisy chain loop interconnection topology.
- Fig. IB is a bidirectional daisy chain interconnection topology.
- Fig. 2 is a functional block diagram of a port transportation unit according to an embodiment of the present invention.
- Fig. 3 is a representative signal timing diagram for the port transportation unit of Fig. 2.
- FIGs. 4A-4D are representations of data in transit from source ports to destination ports, according to system clock snapshots, according to an embodiment of the present invention.
- FIGs. 5A-5D are representations of data in transit from source ports to destination ports, according to system clock snapshots, according to another embodiment of the present invention.
- Fig. 1 shows two examples of four daisy chained ports.
- Fig. IA depicts a ring-like daisy chain in a unidirectional loop.
- Fig. IB depicts an open ended daisy chain configuration, where bidirectional connections make it a bidirectional loop. Note, though, that while the chain flow in Fig. IA is unidirectional, additional chain connections and transportation units could be added to the topology to make bidirectional connections.
- the basic forms of the design embodiments described herein according to the present invention are designated as a self-timed Time Division Multiplexed (TDM) bus with a daisy chain loop configuration, similar to that as depicted in Fig. IA, with wave-front relay self-timing, instead of a local very-high-frequency clock generator.
- TDM Time Division Multiplexed
- Data is self-timed and clocked by a delayed version of one of the self-timing clocks when it arrives at a unit, to provide for a reliable reception. If the data that arrives at the unit needs to be forwarded to a subsequent unit, then the data and the self-timing clocks are re-synchronized before the forwarding operation.
- the transportation unit has three main functions, which are (1) transmission, (2) reception, and (3) relay with data and clock re-synchronization.
- the transportation unit also has a function control block that communicates with the ports, and also loads data to and takes data from the TDM bus at the proper time.
- Fig. 2 depicts a block diagram of the transportation unit.
- the transportation unit consists of clock selection circuits 213 and 227, data path multiplexer 211, control logic block 215, relay data registers 212, a new clock generation block 214, three delay blocks 228, 229, 230, and two selective adjustable delay blocks 218 and 219.
- Fig. 3 gives an example of a signal timing diagram for some of the transportation unit signals, with a time multiplexing factor of four.
- the notation that is used in Fig. 3 includes: Dp0 ⁇ Dp3: Data from the previous port, in an ascending order of when it is sent. Dp2_last, Dp3_last, DpO next: The Dp2 and Dp3 from the last global unit clock cycle, and the DpO for next global unit clock cycle. Dsend: Data to be sent from this transportation unit to the next unit. Dsend last, Dsend next: The Dsend data from the last global unit clock cycle and the Dsend data for the next global unit clock cycle.
- Signal 202 is data to the next unit, which becomes signal 201, data from the previous unit, when it arrives at the next unit.
- signal 201 data from the previous unit
- clocks to next unit becomes signal 204, clocks from previous units, when it arrives at the next unit.
- the data is fed in a multi-bit payload.
- Self-timing clocks have M instances, where M is the number of relays that the units perform within one global unit clock cycle.
- the clock edge is defined as a rising or falling edge of signal
- One clock cycle is defined as the time between two consecutive ones of the selected type of clock edges (rising or falling). At any given time, only one out of M instances of the self-timing clocks are active. Self-timing clocks are generated by block 214, the new clock generation block, through an M bit rotating shift register. The shift register in 214 is reset to 2M- 1 using signal 232 during a system reset event. Then the shift register in 214 shifts at every clock edge of the relay clock signal 203. Shift register outputs from 214 are sent out as signal 207.
- the self-timing clock selection circuit 227 is used to select the currently-active self-timing clock from signal 221 - the delayed version of signal 204. The selection is made through a delayed version of signal 231 , which is sent out by the control logic block 215. The selected self-timing clock then becomes signal 225, the data clock. The data clock 225 is sent to block 215, in which there is a rotating shift register, which is reset to 1 during a system reset event, and which is clocked by the data clock signal 225. The shift register outputs are sent out as signal 231, the delayed version of which is ANDed with signal 221, with the output then ORed to generate the self- timing clock selection outputs signal 225.
- the "on" time of the signal 225 can be adjusted, since the shift register is updated at the clock edge of signal 225, and hence signal 221 is updated.
- signal 221 selects the next active self-timing clock in the queue, which is “off at the time of the selection update, and thereby turns signal 225 "off.”
- the "off time of signal 225 is determined by the timing of the clock edge of the selected active clock. This design self-tracks the required "on" time of signal 225 by observing the shift register state change that is driven by signal 225. Delay that is added by block 228 adds margin to the minimum “on” time signal 225, so that signal 225 meets a robust operation requirement from the flip-flops that are driven by it.
- Signal 225 and the delayed version of signal 208 are selected by multiplex block 213 through a delayed version of signal 226 and a delayed version of an end of relay signal 226 from block 215.
- the output of block 213 becomes signal 203, the relay clock signal.
- Delay 229 is added between the global unit clock signal 208 and the input of block 213 to allow sufficient setup time at the relay data registers 212, when new data comes from the port or logic core on the global unit clock signal 208.
- Fig. 3 depicts the delay for these signals.
- the "on" time of signal 203 is also self-tracked through the state change of the end of relay signal 226, which is driven by the relay clock signal 203.
- Delay 230 adds margin to the "on" time of signal 203 for robustness of the circuit.
- Control logic block 215 controls the timing and data flow of the transportation unit. It sends a selection signal 205 at a proper time to the multiplex block 211 to select between signal 220, the delayed version of signal 201, which is data from the previous unit, and signal 206, data to be sent, to be connected to internal data bus 209. Example timing can be found in Fig. 3. If signal 206 (data to be sent) is selected, then new data is loaded onto the bus, otherwise data from the previous unit is forwarded and made ready for the relay 212.
- the transportation unit design optionally includes an adjustable delay block 218, which can be inserted when the circuit delay in a given design is not sufficiently long so as to guarantee an appropriate hold time.
- the control logic 215 also captures data at the proper time (synchronized to the data clock 225) from signal 220 when data addressed to this unit arrives.
- an optional adjustable delay block 219 can be applied to signal 204 before the selector block 213, if the selected self-timing clock comes too early to guarantee an adequate setup time.
- the timing of loads delivered to the bus - and captures taken from the bus - can be programmable or hardwired.
- the control logic 215 contains a set of counter or shift registers.
- the counter or shift registers reset according to the global unit clock 208.
- the timing of loads and captures are represented in one embodiment as counter values that get compared to counter states, or a set of register bits that are looked up according to the content of the shift register.
- the clock for the counter/shift registers is the relay clock signal 203.
- the above mechanism can also be used to generate the end of relay signal 226.
- Signal 226 is reset to be asserted at a system reset.
- Fig. 3 shows an example of how data is loaded to the bus at the first relay clock edge 203 after the signal 208 clock edge. Data capture is enabled at the fourth relay clock edge 203 and captured at the fourth data clock edge 226. Finally, the end of relay signal 226 is also set at the fourth relay clock edge 203.
- a pipeline architecture can be applied to the design. For example, signal 205 can be generated one local clock earlier than signal 206 can be clocked into the relay registers 212. Fig. 3 shows an example of that.
- Block 215 also contains three sets of registers: transmission data registers 216, receiving data registers 217, and control registers 224.
- the transmission registers 216 supply the data to be loaded on the bus
- the receiving registers 217 are a one level FIFO that store data that is captured from the bus.
- Control registers 224 contain control information.
- the control registers 224 are mostly configuration registers - for example, adjustable delay settings that drive signals 222 and 223, flow control registers that control timing of loads and captures, and so forth.
- the registers are set or read by corresponding port logic elements or a host that communicates with them through the port communication channel 210.
- Each transportation unit along the daisy chain loop starts to transmit data and a sync bit (or bits) at the global unit clock edge.
- Each unit receives data and clocks - self-timing information to use a more generic term - from a previous port according to the loop flow direction, including wire delay between the two ports.
- Each unit then recovers and generates a relay clock from the received sync bits with an adjustable delay.
- the regenerated local relay clock then clocks in the received data and generates new sync bits at the same time, to send to the next port in the ring.
- This design allows transmitted data and sync bits to be relayed to the next port, as well as to be re-synchronized at each transportation unit.
- This relay process continues until the data reaches its destination. In some embodiments, all relay processes are finished within one global clock cycle. The above process then repeat with every global unit clock cycle.
- One example of the relay process is depicted in Figs. 4A-4D, where:
- N Number of ports in the point-to-point communication system
- the number of connections for a direct point-to-point connection is N* (N- 1)* W.
- the number of connections for a TDM point-to-point connection can be calculated as (N-l)*(W+n), as Table 1 shows. So the "wire savings" is calculated as:
- the savings over a direct P2P connection as measured in the total connection length of the interconnects is not calculated here, because it depends significantly on the actual port locations and routing plan.
- the total connection length in a TDM P2P connection can be calculated as given below, assuming that the port to port routing distance is a constant 1:
- Table 1 demonstrates a simple way to establish a P2P connection using a TDM bus, and demonstrates a significant reduction in the number of connections as compared to a direct P2P connection. The number of connections can be further reduced to about half, through bus time sharing.
- a chain designated for data that has port destinations that are far away from the source ports can be paired with a chain whose data destinations are closer to the source ports.
- Table 2 depicts two embodiments of chain pairs that share one bus. The total number of chains reduces from N-I to something within the range of (N-l)/2 to N/2+1, depending on whether N is odd or even, and the pairing scheme used. It is appreciated that there are other sharing schemes that are comprehended within the scope of the present invention that can be used to reduce the total number of interconnects.
- Fig. 5 illustrates how the relay process works within the time share embodiment, with the same definitions for the terms as provided above in regard to Figs. 4.
- t 0 (synchronized to the system clock), and the data departs from the source ports.
- t 1, and the data is in transit.
- Modifications can be made to the transportation unit control block 215 (as depicted in Fig. 2) to facilitate the time share embodiment.
- the control logic can be modified such that it can load and capture more than once within a single global unit clock cycle, at proper timings.
- the capacity of the registers 216 and 217 can be increased according to the chosen time share scheme, and the load and capture can be operated with the proper storage registers in a proper order.
- the worst case delay for the proposed TDM bus which is a result of passing through N-I relay stages, might be too long to meet the speed requirements of the interconnection.
- Using a bi-directional TDM bus for chains that have a large number of relay stages can reduce the worst case number of relay stages from N-I to (N-l)/2 when N is odd, and to N/2 when N is even.
- Table 3 provides an example of such a bi-directional TDM bus.
- bus time sharing technique described in the previous section can also be applied to a bidirectional TDM bus, to reduce the number of connections.
- each port 100 can handle M arbitration requests.
- Each arbitration request goes to each port 100, one of which ports 100 is depicted in Fig. 6. Therefore, there are M*N arbitraion requests in each port 100 and M*N*N total arbitration requests for all of the N ports 100.
- the N ports 100 can be connected as a bidirectional daisy chain as depicted in Fig. IB, or in a unidirectional daisy chain loop or ring as depicted in Fig. IA. In the embodiment depicted in Fig.
- each port 100 includes a clock generator 102 that operates at a frequency that is K times greater than the signal that it receives from the chip clock 110.
- the port 100 also has M banks of shift registers 104, where each of the M banks has N shift registers, which are used for temporary storage.
- the port 100 also includes arbitration request registers 106.
- Each arbitration request is associated with one of the banks 104 of N shift registers.
- Each register in the appropriate bank 104 of N shift registers stores a request that comes from one of the N ports 100.
- the N shift registers contain the corresponding arbitration requests from all of the ports 100.
- signals ArbReqO, ArbReql, ... ArbReqM-1 are arbitration requests that are generated from the arbitration unit, which requests go to the N ports 100 for arbitration.
- the requests are loaded into RegO of the appropriate bank 104 of shift registers at the rising edge of the chip clock 110 signal, when the load data (LD) signal is asserted.
- Control block 108 outputs the load data signal to register RegO of every bank 104 of N shift registers, to initiate the loading of the arbitration requests into the ring structure.
- the load data signal is asserted when the clock generator 102 is disabled, which occurs when the chip clock 110 signal is low, and then the load data signal is de-asserted after the rising edge of the NCLK signal.
- RqtinO, Rqtinl, ... RqtinM-1 are M ring signals that are received from the previous port 100, and are routed to the input of register RegO of every one of the M banks 104 of N shift registers.
- the outputs of RegO (RqtO outO, Rqtl outO, ... RqtM- l outO) are routed to the inputs RqtinO, Rqtinl, .. RqtinM-1 of the next port 100 in the ring or chain.
- the RqtO outO, Rqtl outO, ... RqtM- l outO signals contain the arbitration requests 0 of the corresponding port 100 after the first clock signal from the clock generator 102.
- the clock generator 102 is enabled on the rising edge of the chip clock 110 signal, and continues to be enabled until the internal counter in the control unit 108 reaches a pre-defined number of clock cycles. When the clock generator 102 is disabled, then no more clock signals are generated, and the clock signal NCLK stays low until the next rising edge of the chip clock 110 signal.
- the internal counter in the control unit 108 disables the clock generator 102 when it counts to N and the arbitration requests from each port 100 are shifted to the desired port 100.
- the output of the bank 104 of N shift registers goes to the arbitration request registers 106, and is latched at the next rising of the chip clock 110 signal. Because there are M banks 104 of N shift registers, there are M banks 110 of arbitration request registers. The output of the M banks 110 of arbitration request registers goes to the arbitration unit, to determine the connection between the ports 100.
- the various embodiments of the present invention newly describe a point-to-point TDM bus using a wave-front relay self-timing technique, a new design for a transportation unit, a TDM bus time share technique, and a bidirectional TDM bus.
- the P2P TDM bus described herein significantly reduces the number of connections that are required between ports, as compared to a direct-link point-to- point topology. Specifically, the P2P TDM bus uses O(N) connections, while the direct P2P link bus requires O(N 2 ) connections. Further, the P2P TDM bus significantly reduces the overall routing area that is required for a P2P connection. The P2P TDM bus can reduce the impact of wire delay by increasing wire width and pitch in exchange for a lesser number of interconnections. The P2P TDM bus described herein also reduces interconnect power dissipation due to reduced wire loads.
- the wave-front relay self-timing technique described herein is a very effective technique for the P2P TDM bus. For every global unit clock cycle, all ports on the connected P2P network start by sending data and self-timing information to the next port down the chain. Meanwhile, every port is ready to receive data and self-timing information from the previous port in the chain. Each port uses the self-timing information to re-synchronize and then relay the data, and generates new self-timing information - along with the relayed data - to send to the next port down the chain. Each port extracts the clock signal from the incoming self-timing information, and in doing so removes any need for a high speed (multiple clock rate) clock at each port.
- the wave-front relay self-timing technique described herein also reduces the matching requirement that is imposed by a local high frequency clock generator, and also removes any data/clock mismatching accumulation along the ports that are used for the synchronized relay.
- the wave-front relay self-timing technique also limits the relay process to just one global unit clock cycle, and re-synchronizes all of the relay process at the global clock edge, which prevents path mismatching accumulation from one relay process to the next.
- the wave-front relay self-timing technique increases daisy chain performance, because the relay delay is typically smaller than a local high frequency clock period.
- the transportation unit of the present design features sync detection, local clock generation, data path load and capture functions, and data/sync bit(s) resynchronization, all of which enable the wave-front relay self-timing technique. Further, the transportation unit design as described herein features an additional delay between the unit clock and the start of the relay clock, which allows for the same unit clock cycle data to be used on the bus, and avoids extra clock cycle latency or extra storage.
- the transportation unit includes optional adjustable delays on the data path and the sync path, which allows the hold time and the setup time for relay to be adjusted separately.
- the transportation unit has a separate local clock for data relay and data capture, which allows more time for the data capture operation to finish, thus maximizing the use of the full unit clock cycle for transportation - this also permits a higher speed performance.
- the TDM bus time share technique described herein reduces the number of P2P TDM interconnections by about half, while the bidirectional design also doubles the P2P TDM bus system performance.
- the P2P TDM bus is simplified and modified for a crossbar application, with or without broadcasting.
- the P2P TDM bus can also be cascaded and bridged for multi-stage interconnects. If latency is allowed in a fast system, then the global unit clock speed can be reduce to multi system clock cycles, and the bus can be widened accordingly to meet the throughput requirement.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Small-Scale Networks (AREA)
- Information Transfer Systems (AREA)
- Bus Control (AREA)
- Electronic Switches (AREA)
- Logic Circuits (AREA)
- Time-Division Multiplex Systems (AREA)
- Use Of Switch Circuits For Exchanges And Methods Of Control Of Multiplex Exchanges (AREA)
Priority Applications (7)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/123,124 US8503482B2 (en) | 2008-11-19 | 2008-11-19 | Interconnects using self-timed time-division multiplexed bus |
CN200880129838.3A CN102057360B (zh) | 2008-11-19 | 2008-11-19 | 使用自定时的时分复用总线的互连 |
JP2011536301A JP5731391B2 (ja) | 2008-11-19 | 2008-11-19 | 自己タイミング型時分割多重化バスを使用した相互接続 |
PCT/US2008/083974 WO2010059150A1 (en) | 2008-11-19 | 2008-11-19 | Interconnects using self-timed time-division multiplexed bus |
KR1020117010181A KR101340763B1 (ko) | 2008-11-19 | 2008-11-19 | 신호 송신 방법 및 전송 유닛 |
EP08878325.3A EP2366154A4 (de) | 2008-11-19 | 2008-11-19 | Einen selbstgetimten zeitgemultiplexten bus verwendende verbindungselemente |
TW098102345A TWI426391B (zh) | 2008-11-19 | 2009-01-21 | 傳送訊號的方法及其傳遞單元 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/US2008/083974 WO2010059150A1 (en) | 2008-11-19 | 2008-11-19 | Interconnects using self-timed time-division multiplexed bus |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010059150A1 true WO2010059150A1 (en) | 2010-05-27 |
Family
ID=42198388
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2008/083974 WO2010059150A1 (en) | 2008-11-19 | 2008-11-19 | Interconnects using self-timed time-division multiplexed bus |
Country Status (7)
Country | Link |
---|---|
US (1) | US8503482B2 (de) |
EP (1) | EP2366154A4 (de) |
JP (1) | JP5731391B2 (de) |
KR (1) | KR101340763B1 (de) |
CN (1) | CN102057360B (de) |
TW (1) | TWI426391B (de) |
WO (1) | WO2010059150A1 (de) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8478917B2 (en) | 2010-09-22 | 2013-07-02 | Microsoft Corporation | Automatic addressing protocol for a shared bus |
US8619821B2 (en) | 2011-03-25 | 2013-12-31 | Invensense, Inc. | System, apparatus, and method for time-division multiplexed communication |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5014362B2 (ja) | 2009-02-25 | 2012-08-29 | キヤノン株式会社 | 情報処理装置及びその制御方法、コンピュータプログラム |
US9058273B1 (en) * | 2013-12-20 | 2015-06-16 | International Business Machines Corporation | Frequency determination across an interface of a data processing system |
US10079650B2 (en) | 2015-12-04 | 2018-09-18 | Infineon Technologies Ag | Robust high speed sensor interface for remote sensors |
WO2017111780A1 (en) * | 2015-12-23 | 2017-06-29 | Intel Corporation | Apparatus and method for distribution of congestion information in a switch |
DE102018220302A1 (de) | 2018-11-26 | 2020-05-28 | Festo Ag & Co. Kg | Kommunikationseinheit, Steuergerät, Feldgerät, Kommunikationssystem und Verfahren |
DE102018220301A1 (de) * | 2018-11-26 | 2020-05-28 | Festo Ag & Co. Kg | Kommunikationseinheit, Steuergerät, Kommunikationssystem und Verfahren |
CN114791896A (zh) * | 2021-01-26 | 2022-07-26 | 意法半导体(格勒诺布尔2)公司 | 片上系统中的时域同步 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20010044861A1 (en) * | 2000-05-16 | 2001-11-22 | Yoshikatsu Niwa | Information processing apparatus, information processing method and bridge utilizing the same |
US20020135408A1 (en) * | 2001-01-16 | 2002-09-26 | Chiu Kenny Kok-Hoong | Method and interface for glitch-free clock switching |
US20040039857A1 (en) * | 1994-01-21 | 2004-02-26 | Hitachi, Ltd. | Data transfer control method, and peripheral circuit, data processor and data processing system for the method |
US20040049616A1 (en) * | 1998-08-12 | 2004-03-11 | Robert Dunstan | Communicating with devices over a bus and negotiating the transfer rate over the same |
US20050102125A1 (en) * | 1998-08-31 | 2005-05-12 | Verisity Design, Inc. | Inter-chip communication system |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE3628299A1 (de) * | 1986-08-21 | 1988-02-25 | Licentia Gmbh | Verfahren und anordnung zum signaltechnisch sicheren uebertragen von seriellen daten zwischen vorzugsweise zweikanaligen sicheren rechnern unter verwendung eines doppelringbussystems |
JPH0623621B2 (ja) | 1987-09-08 | 1994-03-30 | 松下電器産業株式会社 | 石油燃焼温風暖房機 |
JPH03265904A (ja) * | 1990-03-16 | 1991-11-27 | Hitachi Ltd | プロセス制御システム |
JP2809811B2 (ja) | 1990-04-27 | 1998-10-15 | 三井化学株式会社 | 三弗化窒素ガスの製造方法 |
US5576702A (en) * | 1995-01-10 | 1996-11-19 | Samoylenko; Stanislav I. | Method and apparatus for fault-tolerant transmission in multi-channel networks |
US5919265A (en) * | 1996-05-28 | 1999-07-06 | Sun Microsystems, Inc. | Source synchronization data transfers without resynchronization penalty |
JPH11167560A (ja) * | 1997-12-03 | 1999-06-22 | Nec Corp | データ転送システム、このシステムに用いるスイッチング回路、アダプタ及びこのシステムを有する集積回路並びにデータ転送方法 |
US6560215B1 (en) * | 1999-08-24 | 2003-05-06 | Motorola, Inc. | Method and apparatus for updating a timing offset in a communication device |
US6975653B2 (en) * | 2001-06-12 | 2005-12-13 | Agilent Technologies, Inc. | Synchronizing clocks across sub-nets |
CN100578510C (zh) * | 2001-08-14 | 2010-01-06 | 韦里希特外观有限公司 | 时序不灵敏及无假信号逻辑设备 |
US6965952B2 (en) * | 2001-10-15 | 2005-11-15 | Intel Corporation | Bus framer |
US6748506B2 (en) * | 2001-10-15 | 2004-06-08 | Intel Corporation | Bus frame protocol |
US20050066097A1 (en) * | 2003-09-04 | 2005-03-24 | Matsushita Electric Industrial Co., Ltd. | Resource management apparatus |
TWI235248B (en) * | 2004-04-28 | 2005-07-01 | Univ Nat Chiao Tung | Self-healing fiber bragg grating sensor system |
JP4503565B2 (ja) | 2006-08-31 | 2010-07-14 | 日本電信電話株式会社 | 伝送送信装置、伝送受信装置、伝送信号送信方法、および伝送信号受信方法 |
CN101055479B (zh) * | 2007-05-29 | 2010-08-11 | 北京中星微电子有限公司 | 一种实现非时钟控制寄存器数据更新的系统及方法 |
US7864747B2 (en) * | 2008-03-28 | 2011-01-04 | Embarq Holdings Company, Llc | System and method for communicating timing to a remote node |
-
2008
- 2008-11-19 US US13/123,124 patent/US8503482B2/en not_active Expired - Fee Related
- 2008-11-19 WO PCT/US2008/083974 patent/WO2010059150A1/en active Application Filing
- 2008-11-19 EP EP08878325.3A patent/EP2366154A4/de not_active Withdrawn
- 2008-11-19 KR KR1020117010181A patent/KR101340763B1/ko not_active IP Right Cessation
- 2008-11-19 CN CN200880129838.3A patent/CN102057360B/zh not_active Expired - Fee Related
- 2008-11-19 JP JP2011536301A patent/JP5731391B2/ja not_active Expired - Fee Related
-
2009
- 2009-01-21 TW TW098102345A patent/TWI426391B/zh not_active IP Right Cessation
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040039857A1 (en) * | 1994-01-21 | 2004-02-26 | Hitachi, Ltd. | Data transfer control method, and peripheral circuit, data processor and data processing system for the method |
US20040049616A1 (en) * | 1998-08-12 | 2004-03-11 | Robert Dunstan | Communicating with devices over a bus and negotiating the transfer rate over the same |
US20050102125A1 (en) * | 1998-08-31 | 2005-05-12 | Verisity Design, Inc. | Inter-chip communication system |
US20010044861A1 (en) * | 2000-05-16 | 2001-11-22 | Yoshikatsu Niwa | Information processing apparatus, information processing method and bridge utilizing the same |
US20020135408A1 (en) * | 2001-01-16 | 2002-09-26 | Chiu Kenny Kok-Hoong | Method and interface for glitch-free clock switching |
Non-Patent Citations (1)
Title |
---|
See also references of EP2366154A4 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8478917B2 (en) | 2010-09-22 | 2013-07-02 | Microsoft Corporation | Automatic addressing protocol for a shared bus |
US8619821B2 (en) | 2011-03-25 | 2013-12-31 | Invensense, Inc. | System, apparatus, and method for time-division multiplexed communication |
Also Published As
Publication number | Publication date |
---|---|
CN102057360A (zh) | 2011-05-11 |
TWI426391B (zh) | 2014-02-11 |
JP5731391B2 (ja) | 2015-06-10 |
US20110211582A1 (en) | 2011-09-01 |
EP2366154A4 (de) | 2016-08-10 |
KR20110095248A (ko) | 2011-08-24 |
CN102057360B (zh) | 2014-01-22 |
JP2012509612A (ja) | 2012-04-19 |
US8503482B2 (en) | 2013-08-06 |
EP2366154A1 (de) | 2011-09-21 |
TW201020798A (en) | 2010-06-01 |
KR101340763B1 (ko) | 2013-12-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8503482B2 (en) | Interconnects using self-timed time-division multiplexed bus | |
US7826460B2 (en) | Network-on-chip apparatus, and method for controlling dynamic frequency for the same | |
CN111971648B (zh) | 异步多时钟域数据流接合及再同步系统和方法 | |
US11196587B2 (en) | Permutated ring network | |
JPH11316706A (ja) | データ高速転送同期システム及びデータ高速転送同期方法 | |
JPH088890A (ja) | 拡張入出力素子 | |
JP2002539525A (ja) | 弾性インターフェース装置およびそのための方法 | |
CN110546922B (zh) | 环形网络 | |
KR100414943B1 (ko) | 콤팩트 피씨아이에 기반한 다중 처리 시스템에서의 클럭분배 장치 및 방법 | |
EP1798887B1 (de) | Isochrone Synchronisierungseinrichtung | |
TWI417741B (zh) | 動態調整通道方向之方法及使用其之晶片網路架構 | |
JP2010500641A (ja) | 電子装置及び通信同期方法 | |
JP5347772B2 (ja) | 転送速度設定方法、データ転送装置及び情報処理システム | |
WO2009110588A1 (ja) | データ転送装置及び方法並びに半導体回路 | |
US7861018B2 (en) | System for transmitting data between transmitter and receiver modules on a channel provided with a flow control link | |
EP2515443A1 (de) | Datenseriellumsetzer | |
US20050094619A1 (en) | Apparatus and method for asynchronously controlling data transfers across long wires | |
EP1952583A2 (de) | System von virtuellen datenkanälen zwischen taktgrenzen in einer integrierten schaltung | |
JP2008090598A (ja) | 記憶装置 | |
JP2007531064A (ja) | 集積回路チップ(パイプライン式の挿入のための回路および方法) | |
Saneei et al. | A mesochronous technique for communication in network on chips | |
JPH11345053A (ja) | 情報処理装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200880129838.3 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 08878325 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011536301 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13123124 Country of ref document: US |
|
ENP | Entry into the national phase |
Ref document number: 20117010181 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
REEP | Request for entry into the european phase |
Ref document number: 2008878325 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008878325 Country of ref document: EP |