US20090327548A1 - Scalable Bus Structure - Google Patents
Scalable Bus Structure Download PDFInfo
- Publication number
- US20090327548A1 US20090327548A1 US12/551,590 US55159009A US2009327548A1 US 20090327548 A1 US20090327548 A1 US 20090327548A1 US 55159009 A US55159009 A US 55159009A US 2009327548 A1 US2009327548 A1 US 2009327548A1
- Authority
- US
- United States
- Prior art keywords
- sub
- channels
- bus
- data
- channel
- 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.)
- Abandoned
Links
Images
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
- G06F13/14—Handling requests for interconnection or transfer
-
- 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
- G06F13/4265—Bus transfer protocol, e.g. handshake; Synchronisation on a point to point 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
-
- 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
- G06F13/40—Bus structure
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
Definitions
- the present disclosure relates generally to digital systems, and more specifically, to a scalable bus structure.
- a bus is a channel or path between components in a computer, a computer subsystem, a computer system, or other electronic system.
- a shared bus provides a means for any number of components to communicate over a common path or channel.
- shared bus technology has been supplemented by point-to-point switching connections. Point-to-point switching connections provide a direct connection between two components on the bus while they are communicating with each other. Multiple direct links may be used to allow several components to communicate at the same time.
- a common configuration for a computer includes a microprocessor with system memory.
- a high bandwidth system bus may be used to support communications between the two.
- Bridges may be used to efficiently transfer data between the higher and lower bandwidth buses, as well as provide the necessary protocol translation. Each of these buses has been implemented with different protocols and may have a wide variation in performance requirements between them.
- a method of communicating over a bus includes transmitting a first data type in a first type field over a first sub-channel of a transmit channel of the bus while concurrently transmitting a second data type in a second type field over a second sub-channel of the transmit channel of the bus. The method also includes receiving data over a receive channel of the bus while transmitting the first data type and the second data type over the transmit channel.
- a method of communicating data includes during a first time period, communicating first data over a first number of sub-channels of a transmit channel to a receiving component. The method also includes during a second time period, communicating second data over a second number of sub-channels of the transmit channel to the receiving component.
- the first number of sub-channels and the second number of sub-channels are independently selectable from each other, where the second number of sub-channels is selectable from a total number of sub-channels available for transmission, and where each sub-channel is operable to carry address information, control signals, and write data.
- a method of communicating data includes determining a first number of sub-channels of a transmit channel that are available for transmission of data from a sending component to a receiving component. The method also includes determining a second number of sub-channels of a receive channel that are available for transmission of data from the receiving component to the sending component. The method also includes sending data concurrently over each of the first number of sub-channels, and receiving data concurrently communicated over each of the second number of sub-channels.
- a method of communicating data includes receiving at a bridge multiple data types from a sending component over a plurality of sub-channels of a first bus, where each sub-channel is operable to carry address information, control signals, and write data.
- the method also includes transmitting the multiple data types from the bridge to a receiving component over a second bus, where the first bus has a first bandwidth and the second bus has a second bandwidth, where the first bandwidth is greater than the second bandwidth.
- a processing system includes a sending component configured, during a first time period, to communicate first data over a first number of sub-channels of a transmit channel to a receiving component and configured, during a second time period, to communicate second data over a second number of sub-channels of the transmit channel to the receiving component.
- the first number of sub-channels and the second number of sub-channels are independently selectable from each other, where the second number of sub-channels is selectable from a total number of sub-channels available for transmission, and where each sub-channel is operable to carry address information, control signals, and write data.
- FIG. 1 is a conceptual block diagram illustrating an example of a point-to-point connection over a two channel bus between two components in a processing system
- FIG. 2 is a timing diagram showing a read and write operation between two components in a processing system having a point-to-point connection over a two channel bus;
- FIG. 3 is a conceptual block diagram illustrating an example of a point-to-point connection over a high performance two channel bus between two components in a processing system
- FIG. 4 is a conceptual block diagram illustrating the time division multiplexed nature of the high performance bus of FIG. 3 ;
- FIG. 5 is a conceptual block diagram illustrating an example of a point-to-point connection over a low bandwidth two channel bus between two components in a processing system
- FIG. 6 is a conceptual block diagram illustrating the time division multiplexed nature of the low bandwidth bus of FIG. 5 ;
- FIG. 7 is a conceptual block diagram illustrating an example of a point-to-point connection between a high performance component and a lower bandwidth component through a bridge.
- the bus may be scalable in terms of width and clock frequency to support the bandwidth requirements of the various components.
- the bus may also use a common architecture and signaling protocol for all scalable configurations. This may be achieved by reducing the signaling protocol of the bus to only those signals necessary to either transmit or receive information.
- the bus may be configured with a “transmit channel” that provides a generic medium for broadcasting information from a sending component to a receiving component using the same signaling protocol in a time division multiplexed fashion.
- a “receive channel” may also use the same signaling protocol to broadcast information from the receiving component to the sending component.
- FIG. 1 is a conceptual block diagram illustrating this fundamental concept.
- a point-to-point connection over a bus between two components is shown in a processing system.
- the processing system 100 may be a collection of components that cooperate to perform one or more processing functions.
- the processing system will be a computer, or resident in a computer, and capable of processing, retrieving and storing information.
- the processing system may be a stand-alone system.
- the processing system may be embedded in any device, including by way of example, a cellular telephone.
- the bus 106 is a dedicated bus between the sending component 102 and the receiving component.
- the sending component 102 communicates with the receiving component 104 with a point-to-point connection over the bus 106 through a bus interconnect (not shown).
- a bus interconnect not shown.
- the sending component 102 may be any type of bus mastering component including, by way of example, a microprocessor, a digital signal processor (DSP), a direct memory access controller, a bridge, a programmable logic component, discrete gate or transistor logic, or any other information processing component.
- DSP digital signal processor
- a bridge a programmable logic component, discrete gate or transistor logic, or any other information processing component.
- the receiving component 104 may be any storage component, including, by way of example, registers, memory, a bridge, or any other component capable of retrieving and storing information.
- the storage capacity at each address location of the receiving component may vary depending on the particular application and the overall design constraints. For the purposes of explanation, the receiving component will be described with a storage capacity of 1-byte per address location.
- the sending component 102 may read from or write to the receiving component 104 .
- the sending component may broadcast an address location, the appropriate control signals, and the payload to the receiving component 104 on the transmit channel 108 .
- the “payload” refers to the data associated with a particular read or write operation, and in this case, a write operation.
- the control signals may include transfer qualifiers.
- transfer qualifier refers to a parameter that describes an attribute of a read operation, a write operation, or another bus related operation.
- the transfer qualifiers may include a “payload size signal” to indicate the number of data bytes contained in the payload. If the payload is multiple bytes, then the receiving component 104 may store the payload in a block of sequential address locations beginning with the address location broadcast on the transmit channel 108 . By way of example, if the sending device 102 broadcasts an address location 100 HEX followed by a 4-byte payload, the receiving component 104 may write the payload to a block of sequential address locations starting at 100 HEX and ending at 103 HEX .
- the control signals may also include write byte enables. “Write byte enables” may be used to indicate which byte lanes on the transmit channel 108 will be used to broadcast the payload for a write operation. By way of example, a 2-byte payload broadcast on an 32-bit transmit channel 108 may use 2 of the 4 byte lanes. The write byte enables may be used to indicate to the receiving component 104 which of the 2 byte lanes on the transmit channel 108 will be used to broadcast the payload.
- the address location and the appropriate transfer qualifiers may be the only information that needs to be broadcast on the transmit channel 108 .
- the transfer qualifiers may include a payload size signal to indicate the number of data bytes contained in the payload.
- the receiving component 104 may acknowledge the broadcast and send the payload on the receiving channel 110 . If the payload is multiple bytes, then the receiving component 104 may read the payload from a block of sequential address locations beginning with the address location broadcast on the transmit channel 108 .
- the receiving component 104 may retrieve the payload from a block of sequential address locations starting at 200 HEX and ending at 203 HEX .
- the sending component 102 has total control of the transmit channel 108 and may broadcast one or more address locations with their associated control signals prior to, during, or after an active write operation. Also, the transmit and receive channels 108 and 110 are totally independent, and thus, the broadcasting of address locations, control signals, and write data by the sending component may coincide with the broadcasting of read data by the receiving component 104 . “Write data” refers to data broadcast by the sending component 102 , and “read data” refers to data read from the receiving component 104 and broadcast on the receiving channel 110 .
- An implicit addressing scheme may be used to control the sequence of read and write data operations on the transmit and receive channels 108 and 110 .
- the sending component 102 initiates multiple write operations by broadcasting a series of address locations with the appropriate control signals on the transmit channel 108 , the sending component 102 will broadcast the payload for each write operation in the same sequence in which the address locations are broadcast.
- the sending component 102 initiates multiple read operations by broadcasting a series of address locations with the appropriate control signals
- the receiving component 104 will retrieve the payload for each read operation in the same sequence in which it receives the address locations.
- Transfer tags may be used as an alternative to this implicit addressing scheme.
- the sending component 102 may assign a transfer tag for each read and write operation.
- the transfer tag may be included in the transfer qualifiers broadcast on the transmit channel 108 .
- the sending component 102 may send the transfer tag with the payload, and the receiving component 104 may use the transfer tag recovered from the transfer qualifiers to identify the payload.
- the receiving component 104 may send the recovered transfer tag with the payload, and the sending component may use the transfer tag to identify the payload.
- bus protocol is being presented to illustrate the inventive aspects of a processing system, with the understanding that such inventive aspects may be used with any suitable protocol.
- the basic signaling protocol for the transmit channel is shown below in Table 1. Those skilled in the art will readily be able to vary and/or add signals to this protocol in the actual implementation of the bus structure described herein.
- the same signaling protocol may be used for the receive channel as shown below in Table 2.
- FIG. 2 is a timing diagram illustrating a read and write operation over a 32-bit transmit channel and a 32-bit receive channel.
- a System Clock 202 may be used to synchronize communications between the sending component and the receiving component.
- the System Clock 202 is shown with eleven clock cycles, with each cycle numbered sequentially for ease of explanation.
- a write operation may be initiated by the sending component during the second clock cycle 203 . This may be achieved by asserting the Valid signal 204 and setting the Type field 206 to signal a broadcast of an address location for a write operation. The address location may also be broadcast over the Transmit Channel 208 to the receiving component. In response to this broadcast, the receiving component stores the address location in its address queue.
- the broadcast of the address location may be followed by a control signal broadcast for the write operation in the third clock cycle 205 .
- the sending component may alert the receiving component of the control signal broadcast by keeping the Valid signal 204 asserted and changing the Type field 206 appropriately.
- the control signal broadcast may include the transfer qualifiers and the write byte enables for the write operation.
- the transfer qualifiers may include a payload size signal indicating an 8-byte payload.
- the write byte enables may indicate that the 8-byte payload will be transmitted on all byte lanes of the Transmit Channel 208 .
- the receiving component may determine from this information that the payload broadcast will be broadcast over two clock cycles.
- the first 4-bytes of the payload for the write operation may be broadcast on the Transmit Channel 208 during the fourth clock cycle 207 .
- the sending component may alert the receiving component of the payload broadcast by keeping the Valid signal 204 asserted and changing the Type field 206 to signal a payload broadcast.
- the receiving component recognizes the write data as the first 4-bytes of the payload based on the implicit addressing scheme discussed earlier.
- the first 4-bytes of the payload may be written to the receiving component.
- the Valid signal 204 and the Type field 206 remains unchanged as the second 4-bytes of the payload is broadcast on the Transmit Channel 208 .
- the receiving component has disserted the Transfer Ack signal 210 indicating that it cannot accept the broadcast.
- the sending component may detect that the Transfer Ack signal 210 is not asserted at the end of this fifth clock cycle 209 , and repeat the broadcast of the second 4-bytes of the payload in the following clock cycle 211 .
- the sending component may continue to broadcast the second 4-bytes of the payload every clock cycle until the sending component detects the assertion of the Transfer Ack signal 210 from the receiving component. In this case, only one repeat broadcast is required.
- the second 4-bytes of the payload may be written to the receiving component in the sixth clock cycle. At the end of the sixth clock cycle 211 , the sending component detects the assertion of the Transfer Ack signal 210 , and determines that the broadcast has been received.
- a read operation may be initiated by the sending component during the seventh clock cycle 213 . This may be achieved by asserting the Valid signal 204 and setting the Type field 206 to signal the broadcast of an address location for a read operation. The address location may then be broadcast over the Transmit Channel 208 to the receiving component. In response to this broadcast, the receiving component stores the address location in its address queue.
- the broadcast of the address location may be followed by a control signal broadcast for the read operation in the eighth clock cycle 215 .
- the sending component may alert the receiving component of the control signal broadcast by keeping the Valid signal 204 asserted and changing the Type field 206 appropriately.
- the control signal broadcast may include the transfer qualifiers for the read operation.
- the transfer qualifiers may include a payload size signal indicating a 4-byte payload.
- the receiving component may determine from this information that the payload broadcast can be broadcast over one clock cycle.
- the receiving component may assert the Valid signal 212 and assert the Type field 214 signaling a payload broadcast on the Receive Channel 216 . Since the Transfer Ack signal 218 is asserted by the sending component, the broadcast of the payload may be completed in one clock cycle. The receiving component detects the assertion of the Transfer Ack signal 218 at the end of the tenth clock cycle 219 , and thereby determines that the broadcast of the payload was successful.
- FIG. 3 is conceptual block diagram illustrating a point-to-point connection between two components over a high performance bus.
- the transmit and receive channels 108 and 110 of the high performance bus may be implemented as multiple sub-channels with each sub-channel being 32-bits wide. In actual implementations, the number of sub-channels and the width of each sub-channel may vary depending on the performance requirements of the particular application.
- the transmit channel includes 4 32-bit sub-channels 108 a - 108 d
- the receive channel includes 2 32-bit sub-channels 110 a - 110 b .
- This implementation may be suitable, by way of example, for a system bus in a computer, or any other high performance bus.
- the term “sub-channel” refers to a group of wires or conductors which may be controlled independently of the other wires or conductors in the channel. This means that each sub-channel may be provided with independent signaling capability.
- This high performance bus may be used by the sending component 102 to simultaneously broadcast several combinations of information.
- the sending component may broadcast a 32-bit address location, 32-bits of control signals including transfer qualifiers and write byte enables, and 8-bytes of write data within a single clock cycle.
- 8-bytes of read data may be broadcast from the receiving component 104 to the sending component 102 within a single clock cycle.
- the various embodiments of the processing system described thus far do not include any other type of information broadcast on the receive channel 110 other than read data, there is no need for sub-channels.
- a single 64-bit receive channel may be implemented to reduce the signaling requirements (i.e., no sub-channels).
- the Type field in the signaling protocol may be extended to allow for the broadcast of other information.
- a “write response” may be broadcast on the receive channel 110 to signal the sending component that the data has been written to the receiving component 104 .
- the write response could be broadcast on the receive channel 110 using one of the reserved Type fields.
- the transmit channel may also be extended to include the broadcast of other types of information that are common in many bus protocols, such as standard commands.
- a microprocessor attached to a bus may need to broadcast information to other components in the system such as a TAB Sync command, or a TAB invalidate command. These commands may be classified in the Type field without the need for additional signaling.
- FIG. 4 is a block diagram illustrating the time division multiplexed nature of a transmit channel 108 with 4 sub-channels 108 a - 108 d .
- a complete 8-byte payload broadcast may be completed across the 4 sub-channels within a single clock cycle. More specifically, during the first clock cycle 401 , the sending component may broadcast a 32-bit address location on the first sub-channel 108 a and 32-bits of control signals on the second sub-channel 108 b for the first write operation. The sending component may also broadcast, during the same clock cycle, the higher order 4-bytes of the payload on the third sub-channel 108 c and the lower order 4-bytes of the payload on the fourth sub-channel 108 d .
- Each sub-channel 108 a - 108 d may be provided with independent signaling capability, and in the case described above, assert the Valid signal with the appropriate Type field for each sub-channel.
- two read operations may be initiated by the sending component during the second clock cycle 403 . This may be achieved by broadcasting a 32-bit address location on the first sub-channel 108 a and 32-bits of control signals on the second sub-channel 108 b for the first read operation, with the appropriate signaling on each sub-channel 108 a - 108 b .
- the sending component may also broadcast a 32-bit address location on the third sub-channel 108 c and 32-bits of control signals on the fourth sub-channel 108 d for the second read operation, again with the appropriate signaling for the sub-channels 108 c - 108 d
- a second write operation and third read operation may be initiated by the sending component during the third clock cycle 405 .
- This may be achieved by broadcasting a 32-bit address location on the first sub-channel 108 a and 32-bits of control signals on the second sub-channel 108 b for the second write operation, with the appropriate signaling on each sub-channel 108 a - 108 b .
- the sending component may also broadcast a 32-bit address location on the third sub-channel 108 c and 32-bits of control signals on the fourth sub-channel 108 d for the third read operation, again with the appropriate signaling for the sub-channels 108 c - 108 d.
- the Transfer Ack signal is asserted on the first and second sub-channels 108 a and 108 b , but not on the third and fourth sub-channels 108 c and 108 d .
- the sending component may detect that the Transfer Ack on the third and fourth sub-channels 108 c and 108 d are not asserted, and thus, determine that the address location and the control signals for the third read operation should be rebroadcast.
- the address location and the control signals for the third read operation are shown being broadcast during the fourth clock 407 on the third and fourth sub-channels 108 c and 108 d , respectively, but may be rebroadcast on any sub-channels during any subsequent clock cycle.
- the receiving component is configured to either accept or reject both the address location and the control signals for the third read operation.
- the receiving component may be configured to accept the address location and reject the control signals, or vice versa, for the same read or write operation.
- the receiving component may be configured to accept or reject the higher or lower order bytes of the payload individually.
- the address for the next read or write operation is not broadcast until the control signals associated with the current read or write operation request is received and acknowledged by the receiving component.
- the sending component may broadcast the payload for the second write operation and attempt for the second time to initiate a third read operation. This may be achieved by broadcasting the higher order 4-bytes of the payload on the first sub-channel 108 a and the lower order 4-bytes of the payload on the second sub-channel 108 b for the second write operation, with the appropriate signaling on each sub-channel 108 a - 108 b .
- the sending component may also rebroadcast the 32-bit address location on the third sub-channel 108 c and 32-bits of control signals on the fourth sub-channel 108 d for the third read operation.
- the ordering of the read/write requests may be implicit by position.
- the sending component may broadcast the first read/write request on the first sub-channel 108 a , the second read/write request on the second sub-channel 108 b , the third read/write request on the third sub-channel 108 c , and the fourth read/write request on the fourth sub-channel 108 d .
- the receiving component may process the requests based on this implicit positioning in order to maintain sequential consistency.
- the receiving component may wait until the data broadcast on the first and second sub-channels 108 a and 108 b during the fourth clock cycle 407 is written to the address location before providing the newly written data at this address location to the receive channel for transmission to the sending component.
- the write data does not need to be broadcast immediately following the broadcast of the write operation request (i.e., the address location and control signals).
- Other higher priority read operation requests and/or commands may be interleaved with the write data broadcast on the transmit channel 108 .
- the sending component if the sending component interleaves the read operation requests and/or commands with the write data, then the sending component should be configured with an address back-off mechanism.
- the sending component samples the Transfer Ack signal 210 following a broadcast on the Transmit Channel 208 . If the sending component fails to detect an asserted Transfer Ack signal 210 , then it may repeat the broadcast during the following clock cycle. The broadcast may be repeated every clock cycle until the sending component detects an asserted Transfer Ack signal 210 .
- a problem may arise when the address queue is full during a read operation request, and therefore, cannot accept any more address locations.
- the receiving component needs to complete the pending write operation in order to free up space in the address queue. In this case, the receiving component is said to be deadlocked.
- the address back-off mechanism is designed to allow the write operation to be completed when the receiving component is in deadlock. This may be achieved by limiting the number of repeat broadcasts by the sending component in connection with a read operation request. If the receiving component does not acknowledge a read operation request with a Transfer Ack signal within a certain number of clock cycles, then the sending component may abort the request by sending the remaining write data in place of the address location for the current read operation request. If there is not a pending write operation that needs to be completed, then the broadcast of the read operation request does not need to be aborted. The broadcast may continue until the receiving component acknowledges the request.
- the address back-off mechanism may not be needed if the sending component does not interleave read operation requests with write data. That is, if the address location for a write operation is followed immediately by the control signals, and then immediately followed by the write data, then the receiving component will never encounter deadlock. However, this may degrade the performance of the receive channel because the sending component may not be able to keep the pipeline of read operations sufficient to fully utilize the bandwidth of the receive channel.
- FIG. 5 is a conceptual block diagram illustrating a point-to-point connection between two components over a low bandwidth bus.
- the low bandwidth bus may be implemented with a single transmit channel 108 and a single receive channel 110 requiring fewer signals and resulting in lower power dissipation.
- the sending component 102 may broadcast information to the receiving component 104 over a 32-bit transmit channel 108
- the receiving component 104 may broadcast information back to the sending component 102 over a 32-bit receive channel 110 .
- this same bus architecture may be implemented with narrower bus widths.
- each read or write operation may now require multiple clock cycles as shown in the block diagram of FIG. 6 .
- two clock cycles are used to initiate a read operation. More specifically, a 32-bit address location may be broadcast on the transmit channel 108 in the first clock cycle 601 , followed by 32-bits of control signals in the following clock cycle 603 .
- a 4-byte payload may be read from the receiving component in response to this request and broadcast on the receive channel 110 in the third clock cycle 605 .
- the sending component may initiate a write operation.
- the write operation uses three clock cycles.
- the sending component broadcasts a 32-bit address location on the transmit channel 108 , followed by 32-bits of control signals in the fourth clock cycle 607 , followed by a 4-byte payload in the fifth clock cycle 609 .
- FIG. 7 is a conceptual block diagram illustrating a point-to-point connection between two components through a bridge.
- the bridge 702 may be used to interface a sending component 102 attached to a high performance bus to a receiving component 104 attached to a lower bandwidth bus.
- the high performance bus may be implemented with a transmit channel 108 having 4 32-bit sub-channels 108 a - 108 d and a receive channel 110 having 2 32-bit receive channels 110 a and 110 b .
- the lower bandwidth bus may be implemented with a single 32-bit transmit channel 108 ′ and a single 32-bit receive channel 110 ′.
- a write operation may be completed between the sending device 102 and the bridge 702 within a single clock cycle using the 4 transmit sub-channels 108 a - 108 d of the high performance bus to broadcast the address location, the control signals, and an 8-byte payload as described earlier in connection with FIGS. 3 and 4 .
- the bridge 702 may buffer and broadcast the information to the receiving component 104 over the 32-bit transmit channel 108 ′ of the lower bandwidth bus in 4 clock cycles as described earlier in connection with FIGS. 5 and 6 .
- an address location and the control signals may be broadcast by the sending component 102 to the bridge 702 on 2 transmit sub-channels of the high performance bus within a single clock cycle.
- the bridge 702 may buffer and broadcast this information to the receiving component 104 over the 32-bit transmit channel 108 ′ in two clock cycles.
- An 8-byte payload may then be broadcast from the receiving component 104 to the bridge 702 on the 32-bit receive channel 110 ′, buffered in the bridge 702 , and then broadcast by the bridge 702 to the sending component 102 on the two receive sub-channels 110 a and 110 b in a single clock cycle.
- DSP digital signal processor
- ASIC application specific integrated circuit
- FPGA field programmable gate array
- a general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine.
- a processor may also be implemented as a combination of computing components, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
- a software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
- a storage medium may be coupled to the processor such that the processor can read information from, and write information to, the storage medium.
- the storage medium may be integral to the processor.
- the processor and the storage medium may reside in an ASIC.
- the ASIC may reside in the sending and/or receiving component, or elsewhere.
- the processor and the storage medium may reside as discrete components in the sending and/or receiving component, or elsewhere.
Abstract
A method of communicating over a bus is disclosed and includes transmitting a first data type in a first type field over a first sub-channel of a transmit channel of the bus while concurrently transmitting a second data type in a second type field over a second sub-channel of the transmit channel of the bus. The method also includes receiving data over a receive channel of the bus while transmitting the first data type and the second data type over the transmit channel.
Description
- This application claims priority from and is a continuation of prior application Ser. No. 11/565,041, filed Nov. 30, 2006, which claims priority from and is a continuation of prior application Ser. No. 10/921,053, filed Aug. 17, 2004, which claims priority from and the benefit of U.S. Provisional Ser. No. 60/542,114, filed Feb. 4, 2004.
- The present disclosure relates generally to digital systems, and more specifically, to a scalable bus structure.
- Computers have revolutionized the electronics industry by enabling sophisticated processing tasks to be performed quickly. These sophisticated tasks may be performed by systems containing a high number of complex components that communicate with one another in a fast and efficient manner using a bus. A bus is a channel or path between components in a computer, a computer subsystem, a computer system, or other electronic system.
- Many buses resident in a computer have traditionally been implemented as shared buses. A shared bus provides a means for any number of components to communicate over a common path or channel. In recent years, shared bus technology has been supplemented by point-to-point switching connections. Point-to-point switching connections provide a direct connection between two components on the bus while they are communicating with each other. Multiple direct links may be used to allow several components to communicate at the same time.
- A common configuration for a computer includes a microprocessor with system memory. A high bandwidth system bus may be used to support communications between the two. In addition, there may also be a peripheral bus which is used to transfer data to peripherals. In some cases, there may also be a configuration bus which is used for the purpose of programming various resources. Bridges may be used to efficiently transfer data between the higher and lower bandwidth buses, as well as provide the necessary protocol translation. Each of these buses has been implemented with different protocols and may have a wide variation in performance requirements between them.
- The use of multiple bus structures in a computer has provided a workable solution for many years. However, as area and power emerge as the major design considerations for integrated circuits, it is becoming increasingly desirable to reduce the complexity of the bus structure.
- In a particular embodiment, a method of communicating over a bus includes transmitting a first data type in a first type field over a first sub-channel of a transmit channel of the bus while concurrently transmitting a second data type in a second type field over a second sub-channel of the transmit channel of the bus. The method also includes receiving data over a receive channel of the bus while transmitting the first data type and the second data type over the transmit channel.
- In another particular embodiment, a method of communicating data includes during a first time period, communicating first data over a first number of sub-channels of a transmit channel to a receiving component. The method also includes during a second time period, communicating second data over a second number of sub-channels of the transmit channel to the receiving component. The first number of sub-channels and the second number of sub-channels are independently selectable from each other, where the second number of sub-channels is selectable from a total number of sub-channels available for transmission, and where each sub-channel is operable to carry address information, control signals, and write data.
- In yet another particular embodiment, a method of communicating data includes determining a first number of sub-channels of a transmit channel that are available for transmission of data from a sending component to a receiving component. The method also includes determining a second number of sub-channels of a receive channel that are available for transmission of data from the receiving component to the sending component. The method also includes sending data concurrently over each of the first number of sub-channels, and receiving data concurrently communicated over each of the second number of sub-channels.
- In another particular embodiment, a method of communicating data includes receiving at a bridge multiple data types from a sending component over a plurality of sub-channels of a first bus, where each sub-channel is operable to carry address information, control signals, and write data. The method also includes transmitting the multiple data types from the bridge to a receiving component over a second bus, where the first bus has a first bandwidth and the second bus has a second bandwidth, where the first bandwidth is greater than the second bandwidth.
- In yet another particular embodiment, a processing system includes a sending component configured, during a first time period, to communicate first data over a first number of sub-channels of a transmit channel to a receiving component and configured, during a second time period, to communicate second data over a second number of sub-channels of the transmit channel to the receiving component. The first number of sub-channels and the second number of sub-channels are independently selectable from each other, where the second number of sub-channels is selectable from a total number of sub-channels available for transmission, and where each sub-channel is operable to carry address information, control signals, and write data.
- It is understood that other embodiments of the present invention will become readily apparent to those skilled in the art from the following detailed description, wherein various embodiments of the invention are shown and described by way of illustration. As will be realized, the invention is capable of other and different embodiments and its several details are capable of modification in various other respects, all without departing from the spirit and scope of the present invention. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.
- Aspects of the present invention are illustrated by way of example, and not by way of limitation, in the accompanying drawings, wherein:
-
FIG. 1 is a conceptual block diagram illustrating an example of a point-to-point connection over a two channel bus between two components in a processing system; -
FIG. 2 is a timing diagram showing a read and write operation between two components in a processing system having a point-to-point connection over a two channel bus; -
FIG. 3 is a conceptual block diagram illustrating an example of a point-to-point connection over a high performance two channel bus between two components in a processing system; -
FIG. 4 is a conceptual block diagram illustrating the time division multiplexed nature of the high performance bus ofFIG. 3 ; -
FIG. 5 is a conceptual block diagram illustrating an example of a point-to-point connection over a low bandwidth two channel bus between two components in a processing system; -
FIG. 6 is a conceptual block diagram illustrating the time division multiplexed nature of the low bandwidth bus ofFIG. 5 ; and -
FIG. 7 is a conceptual block diagram illustrating an example of a point-to-point connection between a high performance component and a lower bandwidth component through a bridge. - The detailed description set forth below in connection with the appended drawings is intended as a description of various embodiments of the present invention and is not intended to represent the only embodiments in which the present invention may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced without these specific details. In some instances, well-known structures and components are shown in block diagram form in order to avoid obscuring the concepts of the present invention. Acronyms and other descriptive terminology may be used merely for convenience and clarity and are not intended to limit the scope of the invention.
- Various components in a processing system may communicate over a bus. The bus may be scalable in terms of width and clock frequency to support the bandwidth requirements of the various components. The bus may also use a common architecture and signaling protocol for all scalable configurations. This may be achieved by reducing the signaling protocol of the bus to only those signals necessary to either transmit or receive information.
- The bus may be configured with a “transmit channel” that provides a generic medium for broadcasting information from a sending component to a receiving component using the same signaling protocol in a time division multiplexed fashion. A “receive channel” may also use the same signaling protocol to broadcast information from the receiving component to the sending component.
-
FIG. 1 is a conceptual block diagram illustrating this fundamental concept. A point-to-point connection over a bus between two components is shown in a processing system. Theprocessing system 100 may be a collection of components that cooperate to perform one or more processing functions. Typically, the processing system will be a computer, or resident in a computer, and capable of processing, retrieving and storing information. The processing system may be a stand-alone system. Alternatively, the processing system may be embedded in any device, including by way of example, a cellular telephone. - In one embodiment of the
processing system 100, thebus 106 is a dedicated bus between the sendingcomponent 102 and the receiving component. In another embodiment of theprocessing system 100, the sendingcomponent 102 communicates with the receivingcomponent 104 with a point-to-point connection over thebus 106 through a bus interconnect (not shown). Moreover, as those skilled in the art will readily appreciate, the inventive aspects described throughout this disclosure are not limited to a dedicated bus or point-to-point switching connection, but may be applied to any type of bus technology including, by way of example, a shared bus. - The sending
component 102 may be any type of bus mastering component including, by way of example, a microprocessor, a digital signal processor (DSP), a direct memory access controller, a bridge, a programmable logic component, discrete gate or transistor logic, or any other information processing component. - The receiving
component 104 may be any storage component, including, by way of example, registers, memory, a bridge, or any other component capable of retrieving and storing information. The storage capacity at each address location of the receiving component may vary depending on the particular application and the overall design constraints. For the purposes of explanation, the receiving component will be described with a storage capacity of 1-byte per address location. - The sending
component 102 may read from or write to the receivingcomponent 104. In the case where the sendingcomponent 102 writes to the receivingcomponent 104, the sending component may broadcast an address location, the appropriate control signals, and the payload to the receivingcomponent 104 on the transmitchannel 108. The “payload” refers to the data associated with a particular read or write operation, and in this case, a write operation. - The control signals may include transfer qualifiers. The term “transfer qualifier” refers to a parameter that describes an attribute of a read operation, a write operation, or another bus related operation. In this case, the transfer qualifiers may include a “payload size signal” to indicate the number of data bytes contained in the payload. If the payload is multiple bytes, then the receiving
component 104 may store the payload in a block of sequential address locations beginning with the address location broadcast on the transmitchannel 108. By way of example, if the sendingdevice 102 broadcasts anaddress location 100 HEX followed by a 4-byte payload, the receivingcomponent 104 may write the payload to a block of sequential address locations starting at 100 HEX and ending at 103 HEX. - The control signals may also include write byte enables. “Write byte enables” may be used to indicate which byte lanes on the transmit
channel 108 will be used to broadcast the payload for a write operation. By way of example, a 2-byte payload broadcast on an 32-bit transmitchannel 108 may use 2 of the 4 byte lanes. The write byte enables may be used to indicate to the receivingcomponent 104 which of the 2 byte lanes on the transmitchannel 108 will be used to broadcast the payload. - In the case where the sending
component 102 reads from the receivingcomponent 104, the address location and the appropriate transfer qualifiers may be the only information that needs to be broadcast on the transmitchannel 108. The transfer qualifiers may include a payload size signal to indicate the number of data bytes contained in the payload. The receivingcomponent 104 may acknowledge the broadcast and send the payload on the receivingchannel 110. If the payload is multiple bytes, then the receivingcomponent 104 may read the payload from a block of sequential address locations beginning with the address location broadcast on the transmitchannel 108. By way of example, if the sendingdevice 102 broadcasts an address location 200 HEX and requests a 4-byte payload, the receivingcomponent 104 may retrieve the payload from a block of sequential address locations starting at 200 HEX and ending at 203 HEX. - In the embodiment of the processing system described thus far, the sending
component 102 has total control of the transmitchannel 108 and may broadcast one or more address locations with their associated control signals prior to, during, or after an active write operation. Also, the transmit and receivechannels component 104. “Write data” refers to data broadcast by the sendingcomponent 102, and “read data” refers to data read from the receivingcomponent 104 and broadcast on the receivingchannel 110. - An implicit addressing scheme may be used to control the sequence of read and write data operations on the transmit and receive
channels component 102 initiates multiple write operations by broadcasting a series of address locations with the appropriate control signals on the transmitchannel 108, the sendingcomponent 102 will broadcast the payload for each write operation in the same sequence in which the address locations are broadcast. Similarly, if the sendingcomponent 102 initiates multiple read operations by broadcasting a series of address locations with the appropriate control signals, the receivingcomponent 104 will retrieve the payload for each read operation in the same sequence in which it receives the address locations. - “Transfer tags” may be used as an alternative to this implicit addressing scheme. The sending
component 102 may assign a transfer tag for each read and write operation. The transfer tag may be included in the transfer qualifiers broadcast on the transmitchannel 108. In the case of a write operation, the sendingcomponent 102 may send the transfer tag with the payload, and the receivingcomponent 104 may use the transfer tag recovered from the transfer qualifiers to identify the payload. In the case of a read operation, the receivingcomponent 104 may send the recovered transfer tag with the payload, and the sending component may use the transfer tag to identify the payload. - The various concepts described thus far may be implemented using any number of protocols. In the detailed description to follow, an example of a bus protocol will be presented. This bus protocol is being presented to illustrate the inventive aspects of a processing system, with the understanding that such inventive aspects may be used with any suitable protocol. The basic signaling protocol for the transmit channel is shown below in Table 1. Those skilled in the art will readily be able to vary and/or add signals to this protocol in the actual implementation of the bus structure described herein.
-
TABLE 1 Signal Definition Driven By Clock the reference clock signal system Valid valid information is being sending component broadcast on the transmit channel Type (2:0) indicates the type of sending component information being broadcast Transfer Ack indicates receiving component receiving component is ready to receive write data Transmit Channel channel driven by the sending sending component component to broadcast information - The same signaling protocol may be used for the receive channel as shown below in Table 2.
-
TABLE 2 Signal Definition Driven By Clock the reference clock signal system Valid valid information is being Receiving component broadcast on the receive channel Type (2:0) Indicates the type of Receiving component information being broadcast Transfer Ack indicates sending component sending component is ready to receive read data Receive Channel channel driven by the Receiving component receiving component to broadcast information - The definition of the Type field used in this signaling protocol is shown in Table 3.
-
TABLE 3 Type Value Definition 000 Reserved 001 Valid Write Address Location 010 Valid Write Control Signals 011 Valid Write Data 100 Reserved 101 Valid Read Address Location 110 Valid Read Control Signals 111 Valid Read Data - The definition of the Valid and Transfer Ack signals in this signaling protocol is shown in Table 4.
-
TABLE 4 Valid; Transfer Ack Definition 0; 0 Valid information is not being broadcast, and the component at the other end is not ready to receive a broadcast 0; 1 Valid information is not being broadcast, but the component at the other end is ready to receive a broadcast 1; 0 Valid information is being broadcast, but the component at the other end is not ready to receive a broadcast 1; 1 Valid information is being broadcast, and the component at the other end is ready to receive a broadcast -
FIG. 2 is a timing diagram illustrating a read and write operation over a 32-bit transmit channel and a 32-bit receive channel. ASystem Clock 202 may be used to synchronize communications between the sending component and the receiving component. TheSystem Clock 202 is shown with eleven clock cycles, with each cycle numbered sequentially for ease of explanation. - A write operation may be initiated by the sending component during the
second clock cycle 203. This may be achieved by asserting theValid signal 204 and setting theType field 206 to signal a broadcast of an address location for a write operation. The address location may also be broadcast over the TransmitChannel 208 to the receiving component. In response to this broadcast, the receiving component stores the address location in its address queue. - The broadcast of the address location may be followed by a control signal broadcast for the write operation in the
third clock cycle 205. The sending component may alert the receiving component of the control signal broadcast by keeping theValid signal 204 asserted and changing theType field 206 appropriately. The control signal broadcast may include the transfer qualifiers and the write byte enables for the write operation. In this case, the transfer qualifiers may include a payload size signal indicating an 8-byte payload. The write byte enables may indicate that the 8-byte payload will be transmitted on all byte lanes of the TransmitChannel 208. The receiving component may determine from this information that the payload broadcast will be broadcast over two clock cycles. - The first 4-bytes of the payload for the write operation may be broadcast on the Transmit
Channel 208 during thefourth clock cycle 207. The sending component may alert the receiving component of the payload broadcast by keeping theValid signal 204 asserted and changing theType field 206 to signal a payload broadcast. In the absence of transfer tags, the receiving component recognizes the write data as the first 4-bytes of the payload based on the implicit addressing scheme discussed earlier. In response to this broadcast, the first 4-bytes of the payload may be written to the receiving component. - In the following
clock cycle 209, theValid signal 204 and theType field 206 remains unchanged as the second 4-bytes of the payload is broadcast on the TransmitChannel 208. However, the receiving component has disserted the Transfer Ack signal 210 indicating that it cannot accept the broadcast. The sending component may detect that theTransfer Ack signal 210 is not asserted at the end of thisfifth clock cycle 209, and repeat the broadcast of the second 4-bytes of the payload in the followingclock cycle 211. The sending component may continue to broadcast the second 4-bytes of the payload every clock cycle until the sending component detects the assertion of the Transfer Ack signal 210 from the receiving component. In this case, only one repeat broadcast is required. The second 4-bytes of the payload may be written to the receiving component in the sixth clock cycle. At the end of thesixth clock cycle 211, the sending component detects the assertion of theTransfer Ack signal 210, and determines that the broadcast has been received. - A read operation may be initiated by the sending component during the
seventh clock cycle 213. This may be achieved by asserting theValid signal 204 and setting theType field 206 to signal the broadcast of an address location for a read operation. The address location may then be broadcast over the TransmitChannel 208 to the receiving component. In response to this broadcast, the receiving component stores the address location in its address queue. - The broadcast of the address location may be followed by a control signal broadcast for the read operation in the
eighth clock cycle 215. The sending component may alert the receiving component of the control signal broadcast by keeping theValid signal 204 asserted and changing theType field 206 appropriately. The control signal broadcast may include the transfer qualifiers for the read operation. In this case, the transfer qualifiers may include a payload size signal indicating a 4-byte payload. The receiving component may determine from this information that the payload broadcast can be broadcast over one clock cycle. - Due to the read latency of the receiving component, a several clock cycle delay may be experienced before the read data is available. Once the 4-byte payload is available, the receiving component may assert the
Valid signal 212 and assert theType field 214 signaling a payload broadcast on the Receive Channel 216. Since theTransfer Ack signal 218 is asserted by the sending component, the broadcast of the payload may be completed in one clock cycle. The receiving component detects the assertion of the Transfer Ack signal 218 at the end of thetenth clock cycle 219, and thereby determines that the broadcast of the payload was successful. -
FIG. 3 is conceptual block diagram illustrating a point-to-point connection between two components over a high performance bus. The transmit and receivechannels bit sub-channels 108 a-108 d, and the receive channel includes 2 32-bit sub-channels 110 a-110 b. This implementation may be suitable, by way of example, for a system bus in a computer, or any other high performance bus. The term “sub-channel” refers to a group of wires or conductors which may be controlled independently of the other wires or conductors in the channel. This means that each sub-channel may be provided with independent signaling capability. - This high performance bus may be used by the sending
component 102 to simultaneously broadcast several combinations of information. By way of example, the sending component may broadcast a 32-bit address location, 32-bits of control signals including transfer qualifiers and write byte enables, and 8-bytes of write data within a single clock cycle. In the case of the receivechannel 110, 8-bytes of read data may be broadcast from the receivingcomponent 104 to the sendingcomponent 102 within a single clock cycle. - Since the various embodiments of the processing system described thus far do not include any other type of information broadcast on the receive
channel 110 other than read data, there is no need for sub-channels. A single 64-bit receive channel may be implemented to reduce the signaling requirements (i.e., no sub-channels). However, in some embodiments of the processing system, the Type field in the signaling protocol may be extended to allow for the broadcast of other information. By way of example, a “write response” may be broadcast on the receivechannel 110 to signal the sending component that the data has been written to the receivingcomponent 104. The write response could be broadcast on the receivechannel 110 using one of the reserved Type fields. In that case, it may be useful to have two independently controlled 32-bit sub-channels so that read data and a write response may be broadcast on the receivechannel 110 simultaneously. With 2 32-bit sub-channels, it may then be possible to simultaneously broadcast 4-bytes of read data, 2-bytes of read data and a 32-bit write response, or 2 32-bit write responses. A single 64-bit receivechannel 110, on the other hand, may be only able to support read data or write responses in any given clock cycle. - In a similar manner, the transmit channel may also be extended to include the broadcast of other types of information that are common in many bus protocols, such as standard commands. By way of example, a microprocessor attached to a bus may need to broadcast information to other components in the system such as a TAB Sync command, or a TAB invalidate command. These commands may be classified in the Type field without the need for additional signaling.
-
FIG. 4 is a block diagram illustrating the time division multiplexed nature of a transmitchannel 108 with 4sub-channels 108 a-108 d. In this example, a complete 8-byte payload broadcast may be completed across the 4 sub-channels within a single clock cycle. More specifically, during thefirst clock cycle 401, the sending component may broadcast a 32-bit address location on thefirst sub-channel 108 a and 32-bits of control signals on thesecond sub-channel 108 b for the first write operation. The sending component may also broadcast, during the same clock cycle, the higher order 4-bytes of the payload on thethird sub-channel 108 c and the lower order 4-bytes of the payload on thefourth sub-channel 108 d. Each sub-channel 108 a-108 d may be provided with independent signaling capability, and in the case described above, assert the Valid signal with the appropriate Type field for each sub-channel. - With the Transfer Ack asserted for each sub-channel 108 a-108 d at the end of the
first clock cycle 401, two read operations may be initiated by the sending component during thesecond clock cycle 403. This may be achieved by broadcasting a 32-bit address location on thefirst sub-channel 108 a and 32-bits of control signals on thesecond sub-channel 108 b for the first read operation, with the appropriate signaling on each sub-channel 108 a-108 b. The sending component may also broadcast a 32-bit address location on thethird sub-channel 108 c and 32-bits of control signals on thefourth sub-channel 108 d for the second read operation, again with the appropriate signaling for the sub-channels 108 c-108 d - With the Transfer Ack asserted for each sub-channel 108 a-108 d at the end of the second clock cycle, a second write operation and third read operation may be initiated by the sending component during the
third clock cycle 405. This may be achieved by broadcasting a 32-bit address location on thefirst sub-channel 108 a and 32-bits of control signals on thesecond sub-channel 108 b for the second write operation, with the appropriate signaling on each sub-channel 108 a-108 b. The sending component may also broadcast a 32-bit address location on thethird sub-channel 108 c and 32-bits of control signals on thefourth sub-channel 108 d for the third read operation, again with the appropriate signaling for the sub-channels 108 c-108 d. - In this example, at the end of the
third clock cycle 405, the Transfer Ack signal is asserted on the first andsecond sub-channels fourth sub-channels fourth sub-channels fourth clock 407 on the third andfourth sub-channels - In the above example, the receiving component is configured to either accept or reject both the address location and the control signals for the third read operation. However, in some embodiments of the processing system, the receiving component may be configured to accept the address location and reject the control signals, or vice versa, for the same read or write operation. Similarly, the receiving component may be configured to accept or reject the higher or lower order bytes of the payload individually. In this case, there needs to be a way to tie a rebroadcast of say the control signals for the third read operation to the address location for the same operation previously broadcast. This may be achieved in a variety of ways. By way of example, once an address location for a read or write operation is sent and acknowledged by the receiving component, the address for the next read or write operation is not broadcast until the control signals associated with the current read or write operation request is received and acknowledged by the receiving component.
- During the
fourth clock cycle 407, the sending component may broadcast the payload for the second write operation and attempt for the second time to initiate a third read operation. This may be achieved by broadcasting the higher order 4-bytes of the payload on thefirst sub-channel 108 a and the lower order 4-bytes of the payload on thesecond sub-channel 108 b for the second write operation, with the appropriate signaling on each sub-channel 108 a-108 b. The sending component may also rebroadcast the 32-bit address location on thethird sub-channel 108 c and 32-bits of control signals on thefourth sub-channel 108 d for the third read operation. - In this high performance bus embodiment, the ordering of the read/write requests may be implicit by position. The sending component may broadcast the first read/write request on the
first sub-channel 108 a, the second read/write request on thesecond sub-channel 108 b, the third read/write request on thethird sub-channel 108 c, and the fourth read/write request on thefourth sub-channel 108 d. The receiving component may process the requests based on this implicit positioning in order to maintain sequential consistency. By way of example, if the address locations for the read and write operations initiated during thethird clock cycle 405 are the same, the receiving component may wait until the data broadcast on the first andsecond sub-channels fourth clock cycle 407 is written to the address location before providing the newly written data at this address location to the receive channel for transmission to the sending component. - In the embodiment of the high performance bus described thus far, the write data does not need to be broadcast immediately following the broadcast of the write operation request (i.e., the address location and control signals). Other higher priority read operation requests and/or commands may be interleaved with the write data broadcast on the transmit
channel 108. However, if the sending component interleaves the read operation requests and/or commands with the write data, then the sending component should be configured with an address back-off mechanism. - As described earlier in connection with
FIG. 2 , the sending component samples the Transfer Ack signal 210 following a broadcast on the TransmitChannel 208. If the sending component fails to detect an assertedTransfer Ack signal 210, then it may repeat the broadcast during the following clock cycle. The broadcast may be repeated every clock cycle until the sending component detects an assertedTransfer Ack signal 210. A problem may arise when the address queue is full during a read operation request, and therefore, cannot accept any more address locations. At the same time, the receiving component needs to complete the pending write operation in order to free up space in the address queue. In this case, the receiving component is said to be deadlocked. - The address back-off mechanism is designed to allow the write operation to be completed when the receiving component is in deadlock. This may be achieved by limiting the number of repeat broadcasts by the sending component in connection with a read operation request. If the receiving component does not acknowledge a read operation request with a Transfer Ack signal within a certain number of clock cycles, then the sending component may abort the request by sending the remaining write data in place of the address location for the current read operation request. If there is not a pending write operation that needs to be completed, then the broadcast of the read operation request does not need to be aborted. The broadcast may continue until the receiving component acknowledges the request.
- The address back-off mechanism may not be needed if the sending component does not interleave read operation requests with write data. That is, if the address location for a write operation is followed immediately by the control signals, and then immediately followed by the write data, then the receiving component will never encounter deadlock. However, this may degrade the performance of the receive channel because the sending component may not be able to keep the pipeline of read operations sufficient to fully utilize the bandwidth of the receive channel.
-
FIG. 5 is a conceptual block diagram illustrating a point-to-point connection between two components over a low bandwidth bus. The low bandwidth bus may be implemented with a single transmitchannel 108 and a single receivechannel 110 requiring fewer signals and resulting in lower power dissipation. In the example shown inFIG. 5 , the sendingcomponent 102 may broadcast information to the receivingcomponent 104 over a 32-bit transmitchannel 108, and the receivingcomponent 104 may broadcast information back to the sendingcomponent 102 over a 32-bit receivechannel 110. Alternatively, this same bus architecture may be implemented with narrower bus widths. - Although this configuration continues to allow for the transmit and receive
channels FIG. 6 . In this example, two clock cycles are used to initiate a read operation. More specifically, a 32-bit address location may be broadcast on the transmitchannel 108 in thefirst clock cycle 601, followed by 32-bits of control signals in the followingclock cycle 603. A 4-byte payload may be read from the receiving component in response to this request and broadcast on the receivechannel 110 in thethird clock cycle 605. - Concurrently with the broadcast of the payload on the receive channel, the sending component may initiate a write operation. In this case, the write operation uses three clock cycles. In the
third clock cycle 605, the sending component broadcasts a 32-bit address location on the transmitchannel 108, followed by 32-bits of control signals in thefourth clock cycle 607, followed by a 4-byte payload in thefifth clock cycle 609. - In many processing systems, some devices may require a high bandwidth interconnect while others can sufficiently operate with a much lower bandwidth interconnect. By using a scalable bus architecture, the implementation of bridges may be implemented with a common signaling protocol.
FIG. 7 is a conceptual block diagram illustrating a point-to-point connection between two components through a bridge. Thebridge 702 may be used to interface a sendingcomponent 102 attached to a high performance bus to areceiving component 104 attached to a lower bandwidth bus. The high performance bus may be implemented with a transmitchannel 108 having 4 32-bit sub-channels 108 a-108 d and a receivechannel 110 having 2 32-bit receivechannels channel 108′ and a single 32-bit receivechannel 110′. - In this example, a write operation may be completed between the sending
device 102 and thebridge 702 within a single clock cycle using the 4 transmitsub-channels 108 a-108 d of the high performance bus to broadcast the address location, the control signals, and an 8-byte payload as described earlier in connection withFIGS. 3 and 4 . Thebridge 702 may buffer and broadcast the information to the receivingcomponent 104 over the 32-bit transmitchannel 108′ of the lower bandwidth bus in 4 clock cycles as described earlier in connection withFIGS. 5 and 6 . - In the case of a read operation, an address location and the control signals may be broadcast by the sending
component 102 to thebridge 702 on 2 transmit sub-channels of the high performance bus within a single clock cycle. Thebridge 702 may buffer and broadcast this information to the receivingcomponent 104 over the 32-bit transmitchannel 108′ in two clock cycles. An 8-byte payload may then be broadcast from the receivingcomponent 104 to thebridge 702 on the 32-bit receivechannel 110′, buffered in thebridge 702, and then broadcast by thebridge 702 to the sendingcomponent 102 on the two receive sub-channels 110 a and 110 b in a single clock cycle. - The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic component, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing components, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
- The methods or algorithms described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. A storage medium may be coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in the sending and/or receiving component, or elsewhere. In the alternative, the processor and the storage medium may reside as discrete components in the sending and/or receiving component, or elsewhere.
- The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (25)
1. A method of communicating over a bus, the method comprising:
transmitting a first data type in a first type field over a first sub-channel of a transmit channel of the bus while concurrently transmitting a second data type in a second type field over a second sub-channel of the transmit channel of the bus; and
receiving data over a receive channel of the bus while transmitting the first data type and the second data type over the transmit channel.
2. The method of claim 1 , wherein transmitting the first data type and the second data type further comprises transmitting address information, control information, or data information, wherein the address information comprises read or write address information, the control information comprises read or write control signals, and the data information comprises write data.
3. The method of claim 1 , further comprising receiving a signal indicative of out of order data.
4. The method of claim 3 , wherein the signal indicative of out of order data includes a transfer tag communicated over the receive channel of the bus.
5. The method of claim 1 , further comprising receiving first return data over a first sub-channel of a receive channel of the bus and receiving second return data over a second sub-channel of the receive channel.
6. The method of claim 1 , further comprising concurrently receiving data over the receive channel of the bus and transmitting multiple data types over at least one of the first sub-channel of the transmit channel of the bus and the second sub-channel of the transmit channel of the bus.
7. The method of claim 1 , further comprising receiving a payload over the receive channel and receiving a transfer tag that identifies the payload over the receive channel.
8. The method of claim 1 , further comprising sending a third data type over a third sub-channel of the transmit channel.
9. A method of communicating data, the method comprising:
during a first time period, communicating first data over a first number of sub-channels of a transmit channel to a receiving component; and
during a second time period, communicating second data over a second number of sub-channels of the transmit channel to the receiving component;
wherein the first number of sub-channels and the second number of sub-channels are independently selectable from each other, wherein the second number of sub-channels is selectable from a total number of sub-channels available for transmission, and wherein each sub-channel is operable to carry address information, control signals, and write data.
10. The method of claim 9 , wherein the first number of sub-channels is different than the second number of sub-channels.
11. The method of claim 9 , wherein the second number of sub-channels is selectable based on performance requirements of an application.
12. The method of claim 9 , wherein a first width of each of the first number of sub-channels is different than a second width of each of the second number of sub-channels.
13. A method of communicating data, the method comprising:
determining a first number of sub-channels of a transmit channel that are available for transmission of data from a sending component to a receiving component;
determining a second number of sub-channels of a receive channel that are available for transmission of data from the receiving component to the sending component;
sending data concurrently over each of the first number of sub-channels; and
receiving data concurrently communicated over each of the second number of sub-channels.
14. The method of claim 13 , further comprising dynamically changing the first number of sub-channels or the second number of sub-channels in response to a traffic condition of the transmit channel.
15. A method of communicating data, the method comprising:
receiving at a bridge multiple data types from a sending component over a plurality of sub-channels of a first bus, wherein each sub-channel is operable to carry address information, control signals, and write data; and
transmitting the multiple data types from the bridge to a receiving component over a second bus, wherein the first bus has a first bandwidth and the second bus has a second bandwidth, wherein the first bandwidth is greater than the second bandwidth.
16. The method of claim 15 , wherein each of the first bus and the second bus is implemented with a common protocol.
17. The method of claim 15 , further comprising:
receiving read data from the receiving component at the bridge over the second bus; and
sending the read data from the bridge to the sending component over the first bus.
18. The method of claim 17 , wherein the second bus includes a plurality of sub-channels.
19. A processing system comprising:
a sending component configured, during a first time period, to communicate first data over a first number of sub-channels of a transmit channel to a receiving component and configured, during a second time period, to communicate second data over a second number of sub-channels of the transmit channel to the receiving component;
wherein the first number of sub-channels and the second number of sub-channels are independently selectable from each other, wherein the second number of sub-channels is selectable from a total number of sub-channels available for transmission, and wherein each sub-channel is operable to carry address information, control signals, and write data.
20. The processing system of claim 19 , wherein the first number of sub-channels is different than the second number of sub-channels.
21. The processing system of claim 19 , wherein the second number of sub-channels is selectable based on performance requirements of an application.
22. The processing system of claim 19 , wherein a first width of each of the first number of sub-channels is different than a second width of each of the second number of sub-channels.
23. An apparatus comprising:
means for transmitting multiple data types from a sending component to a bridge over a plurality of sub-channels of a first bus, wherein each sub-channel is operable to carry address information, control signals, and write data;
means for transmitting the multiple data types from the bridge to a receiving component over a second bus, wherein the first bus has a first bandwidth and the second bus has a second bandwidth, wherein the first bandwidth is greater than the second bandwidth;
means for sending read data from the receiving component to the bridge over the second bus; and
means for sending the read data from the bridge to the sending component over the first bus.
24. The apparatus of claim 23 , wherein each of the first bus and the second bus is implemented with a common protocol.
25. The apparatus of claim 23 , wherein the second bus comprises a plurality of sub-channels.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/551,590 US20090327548A1 (en) | 2004-02-04 | 2009-09-01 | Scalable Bus Structure |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US54211404P | 2004-02-04 | 2004-02-04 | |
US10/921,053 US7209998B2 (en) | 2004-02-04 | 2004-08-17 | Scalable bus structure |
US11/565,041 US7913021B2 (en) | 2004-02-04 | 2006-11-30 | Scalable bus structure |
US12/551,590 US20090327548A1 (en) | 2004-02-04 | 2009-09-01 | Scalable Bus Structure |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/565,041 Continuation US7913021B2 (en) | 2004-02-04 | 2006-11-30 | Scalable bus structure |
Publications (1)
Publication Number | Publication Date |
---|---|
US20090327548A1 true US20090327548A1 (en) | 2009-12-31 |
Family
ID=34811473
Family Applications (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/921,053 Active 2025-03-13 US7209998B2 (en) | 2004-02-04 | 2004-08-17 | Scalable bus structure |
US11/565,041 Active US7913021B2 (en) | 2004-02-04 | 2006-11-30 | Scalable bus structure |
US12/551,590 Abandoned US20090327548A1 (en) | 2004-02-04 | 2009-09-01 | Scalable Bus Structure |
Family Applications Before (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/921,053 Active 2025-03-13 US7209998B2 (en) | 2004-02-04 | 2004-08-17 | Scalable bus structure |
US11/565,041 Active US7913021B2 (en) | 2004-02-04 | 2006-11-30 | Scalable bus structure |
Country Status (12)
Country | Link |
---|---|
US (3) | US7209998B2 (en) |
EP (2) | EP2163993B1 (en) |
JP (2) | JP4685800B2 (en) |
KR (1) | KR100876575B1 (en) |
CN (2) | CN100481053C (en) |
AT (2) | ATE526634T1 (en) |
DE (1) | DE602005022829D1 (en) |
ES (1) | ES2348582T3 (en) |
HK (1) | HK1099823A1 (en) |
IL (3) | IL177240A (en) |
PL (1) | PL1714218T3 (en) |
WO (1) | WO2005078594A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3975018A4 (en) * | 2019-06-21 | 2022-10-19 | Huawei Technologies Co., Ltd. | Integrated chip and data processing method |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7209998B2 (en) * | 2004-02-04 | 2007-04-24 | Qualcomm Incorporated | Scalable bus structure |
US7617343B2 (en) * | 2005-03-02 | 2009-11-10 | Qualcomm Incorporated | Scalable bus structure |
US8107492B2 (en) * | 2006-02-24 | 2012-01-31 | Qualcomm Incorporated | Cooperative writes over the address channel of a bus |
US8108563B2 (en) | 2006-02-24 | 2012-01-31 | Qualcomm Incorporated | Auxiliary writes over address channel |
US7822903B2 (en) * | 2006-02-24 | 2010-10-26 | Qualcomm Incorporated | Single bus command having transfer information for transferring data in a processing system |
EP2442131A1 (en) | 2006-04-10 | 2012-04-18 | Elta Systems Ltd. | Distributed jammer system |
JP5166922B2 (en) * | 2008-03-10 | 2013-03-21 | キヤノン株式会社 | Shared bus control device and control method thereof |
KR20100063219A (en) * | 2008-12-03 | 2010-06-11 | 삼성전자주식회사 | Apparatus and method for writing of bitwise data in bus system |
US9768945B2 (en) | 2012-04-23 | 2017-09-19 | Analog Devices, Inc. | Isolated system data communication |
US9972196B2 (en) * | 2012-04-23 | 2018-05-15 | Analog Devices, Inc. | Isolator system with status data integrated with measurement data |
US9280503B2 (en) * | 2013-04-12 | 2016-03-08 | Apple Inc. | Round robin arbiter handling slow transaction sources and preventing block |
WO2015099719A1 (en) * | 2013-12-26 | 2015-07-02 | Intel Corporation | Multichip package link |
US11086803B2 (en) * | 2018-10-05 | 2021-08-10 | Micron Technology, Inc. | Dynamically configuring transmission lines of a bus |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5065396A (en) * | 1990-01-02 | 1991-11-12 | At&T Bell Laboratories | Inverse multiplexer and demultiplexer techniques |
US5870406A (en) * | 1997-02-19 | 1999-02-09 | Ericsson Inc. | Automatic repeat request(ARQ) data communications method and apparatus |
US5944805A (en) * | 1997-08-21 | 1999-08-31 | Advanced Micro Devices, Inc. | System and method for transmitting data upon an address portion of a computer system bus during periods of maximum utilization of a data portion of the bus |
US6629166B1 (en) * | 2000-06-29 | 2003-09-30 | Intel Corporation | Methods and systems for efficient connection of I/O devices to a channel-based switched fabric |
US20040068603A1 (en) * | 2002-10-03 | 2004-04-08 | International Business Machines Corporation | Reordering of requests between an initiator and the request queue of a bus controller |
US6941256B1 (en) * | 1999-07-30 | 2005-09-06 | Matsushita Electric Industrial Co., Ltd. | Bus structure, database and method of designing interface |
US7136953B1 (en) * | 2003-05-07 | 2006-11-14 | Nvidia Corporation | Apparatus, system, and method for bus link width optimization |
US7782325B2 (en) * | 2003-10-22 | 2010-08-24 | Alienware Labs Corporation | Motherboard for supporting multiple graphics cards |
Family Cites Families (39)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS63146148A (en) * | 1986-12-09 | 1988-06-18 | Nec Corp | Bus system |
US5161162A (en) * | 1990-04-12 | 1992-11-03 | Sun Microsystems, Inc. | Method and apparatus for system bus testability through loopback |
US5418785A (en) * | 1992-06-04 | 1995-05-23 | Gte Laboratories Incorporated | Multiple-channel token ring network with single optical fiber utilizing subcarrier multiplexing with a dedicated control channel |
US5303227A (en) * | 1992-08-03 | 1994-04-12 | Motorola, Inc. | Method and apparatus for enhanced modes in SPI communication |
US5450547A (en) | 1992-10-01 | 1995-09-12 | Xerox Corporation | Bus interface using pending channel information stored in single circular queue for controlling channels of data transfer within multiple FIFO devices |
US5699540A (en) * | 1992-11-16 | 1997-12-16 | Intel Corporation | Pseudo-concurrent access to a cached shared resource |
JPH06259343A (en) * | 1993-03-10 | 1994-09-16 | Hitachi Ltd | Multiple bus control method and system using the same |
US5752076A (en) * | 1995-08-31 | 1998-05-12 | Intel Corporation | Dynamic programming of bus master channels by intelligent peripheral devices using communication packets |
US6209071B1 (en) * | 1996-05-07 | 2001-03-27 | Rambus Inc. | Asynchronous request/synchronous data dynamic random access memory |
US5812878A (en) * | 1996-05-10 | 1998-09-22 | Apple Computer, Inc. | System for DMA transfer wherein controller waits before execution of next instruction until a counter counts down from a value loaded by said controller |
US5896513A (en) * | 1996-07-03 | 1999-04-20 | Intel Corporation | Computer system providing a universal architecture adaptive to a variety of processor types and bus protocols |
US5926831A (en) * | 1996-10-11 | 1999-07-20 | International Business Machines Corporation | Methods and apparatus for control of speculative memory accesses |
US5925118A (en) * | 1996-10-11 | 1999-07-20 | International Business Machines Corporation | Methods and architectures for overlapped read and write operations |
JPH10289202A (en) * | 1997-04-16 | 1998-10-27 | Seiko Epson Corp | Microcomputer circuit |
US5991841A (en) * | 1997-09-24 | 1999-11-23 | Intel Corporation | Memory transactions on a low pin count bus |
US6081860A (en) * | 1997-11-20 | 2000-06-27 | International Business Machines Corporation | Address pipelining for data transfers |
US6292705B1 (en) * | 1998-09-29 | 2001-09-18 | Conexant Systems, Inc. | Method and apparatus for address transfers, system serialization, and centralized cache and transaction control, in a symetric multiprocessor system |
US6167475A (en) * | 1998-07-06 | 2000-12-26 | International Business Machines Corporation | Data transfer method/engine for pipelining shared memory bus accesses |
US6430641B1 (en) * | 1999-05-04 | 2002-08-06 | International Business Machines Corporation | Methods, arbiters, and computer program products that can improve the performance of a pipelined dual bus data processing system |
US6427193B1 (en) | 1999-05-18 | 2002-07-30 | Advanced Micro Devices, Inc. | Deadlock avoidance using exponential backoff |
JP2001092662A (en) * | 1999-09-22 | 2001-04-06 | Toshiba Corp | Processor core and processor using the same |
CN1129071C (en) | 1999-10-27 | 2003-11-26 | 盖内蒂克瓦尔有限公司 | Tunnel transmission structure between components and transmission method |
JP2001154981A (en) * | 1999-11-12 | 2001-06-08 | Geneticware Corp Ltd | Method and device for channel communication between elements |
US6628166B2 (en) * | 2000-03-03 | 2003-09-30 | Tripath Technology, Inc. | RF communication system using an RF digital amplifier |
JP2001265711A (en) * | 2000-03-17 | 2001-09-28 | Casio Comput Co Ltd | Device for transferring data and bus system |
JP3714104B2 (en) * | 2000-03-31 | 2005-11-09 | セイコーエプソン株式会社 | An image processing controller for an electronic printing apparatus and an electronic printing apparatus having the same. |
US6654836B1 (en) * | 2000-06-20 | 2003-11-25 | International Business Machines Corporation | Dual master device for improved utilization of a processor local bus |
US6772254B2 (en) * | 2000-06-21 | 2004-08-03 | International Business Machines Corporation | Multi-master computer system with overlapped read and write operations and scalable address pipelining |
US6594712B1 (en) | 2000-10-20 | 2003-07-15 | Banderacom, Inc. | Inifiniband channel adapter for performing direct DMA between PCI bus and inifiniband link |
US7076595B1 (en) * | 2001-05-18 | 2006-07-11 | Xilinx, Inc. | Programmable logic device including programmable interface core and central processing unit |
US6842816B1 (en) * | 2001-07-31 | 2005-01-11 | Network Elements, Inc. | Configurable glueless microprocessor interface |
US7191271B2 (en) * | 2001-09-20 | 2007-03-13 | Lockheed Martin Corporation | Two level multi-tier system bus |
US20030112805A1 (en) | 2001-12-19 | 2003-06-19 | Stanton Kevin B. | Method and apparatus for providing device-to-device connectivity using shared infiniband NIC device |
JP4314460B2 (en) * | 2003-02-20 | 2009-08-19 | ソニー株式会社 | Data transfer method, data transfer device, and data transfer system |
WO2005071556A1 (en) * | 2004-01-22 | 2005-08-04 | Qualcomm Incorporated | A two channel bus structure to support address information, data, and transfer qualifiers |
US20050182884A1 (en) * | 2004-01-22 | 2005-08-18 | Hofmann Richard G. | Multiple address two channel bus structure |
US7209998B2 (en) * | 2004-02-04 | 2007-04-24 | Qualcomm Incorporated | Scalable bus structure |
US8028143B2 (en) * | 2004-08-27 | 2011-09-27 | Qualcomm Incorporated | Method and apparatus for transmitting memory pre-fetch commands on a bus |
US7185123B2 (en) * | 2004-09-15 | 2007-02-27 | Qualcomm Incorporated | Method and apparatus for allocating bandwidth on a transmit channel of a bus |
-
2004
- 2004-08-17 US US10/921,053 patent/US7209998B2/en active Active
-
2005
- 2005-02-03 AT AT09177248T patent/ATE526634T1/en not_active IP Right Cessation
- 2005-02-03 DE DE602005022829T patent/DE602005022829D1/en active Active
- 2005-02-03 EP EP09177248A patent/EP2163993B1/en not_active Not-in-force
- 2005-02-03 EP EP05722790A patent/EP1714218B9/en not_active Not-in-force
- 2005-02-03 CN CNB2005800089090A patent/CN100481053C/en not_active Expired - Fee Related
- 2005-02-03 WO PCT/US2005/003789 patent/WO2005078594A1/en active Application Filing
- 2005-02-03 JP JP2006552321A patent/JP4685800B2/en not_active Expired - Fee Related
- 2005-02-03 CN CN200910009589.0A patent/CN101493805B/en not_active Expired - Fee Related
- 2005-02-03 ES ES05722790T patent/ES2348582T3/en active Active
- 2005-02-03 PL PL05722790T patent/PL1714218T3/en unknown
- 2005-02-03 KR KR1020067018007A patent/KR100876575B1/en not_active IP Right Cessation
- 2005-02-03 AT AT05722790T patent/ATE477542T1/en not_active IP Right Cessation
-
2006
- 2006-08-02 IL IL177240A patent/IL177240A/en not_active IP Right Cessation
- 2006-11-30 US US11/565,041 patent/US7913021B2/en active Active
-
2007
- 2007-07-09 HK HK07107305.8A patent/HK1099823A1/en not_active IP Right Cessation
-
2009
- 2009-09-01 US US12/551,590 patent/US20090327548A1/en not_active Abandoned
-
2010
- 2010-09-08 JP JP2010200833A patent/JP5180268B2/en not_active Expired - Fee Related
- 2010-11-11 IL IL209279A patent/IL209279A0/en unknown
- 2010-11-11 IL IL209278A patent/IL209278A0/en not_active IP Right Cessation
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5065396A (en) * | 1990-01-02 | 1991-11-12 | At&T Bell Laboratories | Inverse multiplexer and demultiplexer techniques |
US5870406A (en) * | 1997-02-19 | 1999-02-09 | Ericsson Inc. | Automatic repeat request(ARQ) data communications method and apparatus |
US5944805A (en) * | 1997-08-21 | 1999-08-31 | Advanced Micro Devices, Inc. | System and method for transmitting data upon an address portion of a computer system bus during periods of maximum utilization of a data portion of the bus |
US6941256B1 (en) * | 1999-07-30 | 2005-09-06 | Matsushita Electric Industrial Co., Ltd. | Bus structure, database and method of designing interface |
US6629166B1 (en) * | 2000-06-29 | 2003-09-30 | Intel Corporation | Methods and systems for efficient connection of I/O devices to a channel-based switched fabric |
US20040068603A1 (en) * | 2002-10-03 | 2004-04-08 | International Business Machines Corporation | Reordering of requests between an initiator and the request queue of a bus controller |
US7136953B1 (en) * | 2003-05-07 | 2006-11-14 | Nvidia Corporation | Apparatus, system, and method for bus link width optimization |
US7782325B2 (en) * | 2003-10-22 | 2010-08-24 | Alienware Labs Corporation | Motherboard for supporting multiple graphics cards |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3975018A4 (en) * | 2019-06-21 | 2022-10-19 | Huawei Technologies Co., Ltd. | Integrated chip and data processing method |
Also Published As
Publication number | Publication date |
---|---|
ATE526634T1 (en) | 2011-10-15 |
JP2007520832A (en) | 2007-07-26 |
JP5180268B2 (en) | 2013-04-10 |
HK1099823A1 (en) | 2007-08-24 |
US20050172063A1 (en) | 2005-08-04 |
CN101493805A (en) | 2009-07-29 |
DE602005022829D1 (en) | 2010-09-23 |
WO2005078594A1 (en) | 2005-08-25 |
IL177240A0 (en) | 2006-12-10 |
CN101493805B (en) | 2015-07-15 |
CN100481053C (en) | 2009-04-22 |
EP2163993B1 (en) | 2011-09-28 |
US20070088894A1 (en) | 2007-04-19 |
EP2163993A1 (en) | 2010-03-17 |
CN1934557A (en) | 2007-03-21 |
ATE477542T1 (en) | 2010-08-15 |
JP2011044154A (en) | 2011-03-03 |
PL1714218T3 (en) | 2011-02-28 |
KR20060120272A (en) | 2006-11-24 |
IL209278A0 (en) | 2011-01-31 |
ES2348582T3 (en) | 2010-12-09 |
EP1714218B1 (en) | 2010-08-11 |
EP1714218B9 (en) | 2012-03-14 |
EP1714218A1 (en) | 2006-10-25 |
JP4685800B2 (en) | 2011-05-18 |
IL209279A0 (en) | 2011-01-31 |
US7209998B2 (en) | 2007-04-24 |
KR100876575B1 (en) | 2008-12-31 |
IL177240A (en) | 2011-01-31 |
US7913021B2 (en) | 2011-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7913021B2 (en) | Scalable bus structure | |
EP1709543B1 (en) | A multiple address two channel bus structure | |
US7739425B2 (en) | Two channel computer bus architecture | |
JP5231612B2 (en) | Method and apparatus for transmitting a memory prefetch command on a bus | |
KR100874073B1 (en) | Method and apparatus for allocating bandwidth on a transmit channel of a bus | |
US20120096202A1 (en) | Auxiliary writes over address channel | |
US7617343B2 (en) | Scalable bus structure | |
MXPA06008913A (en) | Scalable bus structure | |
MXPA06008326A (en) | A two channel bus structure to support address information, data, and transfer qualifiers | |
JP2000099432A (en) | Serial input/output control circuit |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: QUALCOMM INCORPORATED, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HOFMANN, RICHARD GERARD;SCHAFFER, MARK MICHAEL;REEL/FRAME:023182/0755 Effective date: 20040817 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- AFTER EXAMINER'S ANSWER OR BOARD OF APPEALS DECISION |