US20030229676A1 - Command to transfer data from node state agent to memory bridge - Google Patents
Command to transfer data from node state agent to memory bridge Download PDFInfo
- Publication number
- US20030229676A1 US20030229676A1 US10/413,916 US41391603A US2003229676A1 US 20030229676 A1 US20030229676 A1 US 20030229676A1 US 41391603 A US41391603 A US 41391603A US 2003229676 A1 US2003229676 A1 US 2003229676A1
- Authority
- US
- United States
- Prior art keywords
- node
- agent
- transaction
- cache
- block
- 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
- 238000012546 transfer Methods 0.000 title claims abstract description 40
- 230000001427 coherent effect Effects 0.000 claims description 79
- 230000004044 response Effects 0.000 claims description 78
- 238000000034 method Methods 0.000 claims description 14
- 230000000977 initiatory effect Effects 0.000 claims description 12
- 239000003795 chemical substances by application Substances 0.000 description 126
- 239000000523 sample Substances 0.000 description 25
- 238000004891 communication Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 11
- 238000003066 decision tree Methods 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000003111 delayed effect Effects 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- 238000001693 membrane extraction with a sorbent interface Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000011664 signaling Effects 0.000 description 3
- 238000003786 synthesis reaction Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000003999 initiator Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 102000012199 E3 ubiquitin-protein ligase Mdm2 Human genes 0.000 description 1
- 108050002772 E3 ubiquitin-protein ligase Mdm2 Proteins 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012508 change request Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
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
- G06F13/38—Information transfer, e.g. on bus
- G06F13/40—Bus structure
- G06F13/4004—Coupling between buses
- G06F13/4027—Coupling between buses using bus bridges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
- G06F12/0815—Cache consistency protocols
- G06F12/0817—Cache consistency protocols using directory methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
- G06F12/0815—Cache consistency protocols
- G06F12/0817—Cache consistency protocols using directory methods
- G06F12/082—Associative directories
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30003—Arrangements for executing specific machine instructions
- G06F9/3004—Arrangements for executing specific machine instructions to perform operations on memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30003—Arrangements for executing specific machine instructions
- G06F9/30072—Arrangements for executing specific machine instructions to perform conditional operations, e.g. using predicates or guards
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/30003—Arrangements for executing specific machine instructions
- G06F9/30076—Arrangements for executing specific machine instructions to perform miscellaneous control operations, e.g. NOP
- G06F9/30087—Synchronisation or serialisation instructions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
- G06F2212/10—Providing a specific technical effect
- G06F2212/1048—Scalability
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
- G06F2212/25—Using a specific main memory architecture
- G06F2212/254—Distributed memory
- G06F2212/2542—Non-uniform memory access [NUMA] architecture
Definitions
- This invention is related to coherent memory systems, including coherent distributed memory systems such as cache-coherent nonuniform memory access (CC-NUMA) memory systems.
- coherent distributed memory systems such as cache-coherent nonuniform memory access (CC-NUMA) memory systems.
- CC-NUMA cache-coherent nonuniform memory access
- Memory systems are often designed to be coherent. That is, even though multiple copies of data from a given memory location may exist in the memory system, a read of that memory location returns the most recent data written to that memory location. The most recent data written to that memory location may be determined via an order of accesses to the memory location established according to the coherency mechanism.
- a coherent system may include one or more coherent agents and a memory controller coupled via an interconnect of some kind.
- One mechanism for scaling coherent systems to larger numbers of coherent agents is a distributed memory system.
- memory is distributed among various nodes (which may also include coherent agents), and the nodes are interconnected.
- a coherent agent in one node may access memory in another node.
- One class of techniques for maintaining coherency in a distributed memory system is referred to as cache-coherent, nonuniform memory access (CC-NUMA).
- CC-NUMA cache-coherent, nonuniform memory access
- access to memory may have a varying latency (e.g. memory in the same node as an agent may be accessed more rapidly than memory in another node, and accesses to different nodes may have varying latencies as well), but coherency is maintained. Data from another node may be cached in a given node.
- an agent in a node may be desirable to designate an agent in a node to maintain the state of data transferred to the node from another node. If that agent is to discard the data, a mechanism is desired to ensure that the data is deleted from other agents in the node. Additionally, the mechanism may be required to cause the most recently updated copy of the data, wherever it may reside within the node, to be returned to the node from which the data was read.
- a node comprises a first agent, a second agent, and a third agent, all coupled to an interconnect.
- the first agent is configured to initiate a transaction on the interconnect to transfer a coherency block to the second agent.
- the third agent is configured to transmit the coherency block on the interconnect during a data portion of the transaction instead of the first agent responsive to a state of the coherency block in the third agent.
- the first agent may be designated to store the node state of a remote cache block, and the second agent may be responsible for internode coherency within the node.
- a node in another embodiment, includes an interconnect and a cache coupled to the interconnect.
- the cache is configured to evict a cache block stored therein.
- the cache is configured to initiate a first transaction to transfer the cache block on the interconnect responsive to the cache block being a local cache block, and is configured to initiate a second transaction to transfer the cache block on the interconnect responsive to the cache block being a remote cache block.
- the second transaction is different from the first transaction.
- FIG. 1 is a block diagram of one embodiment of a node.
- FIG. 2 is a block diagram of one embodiment of a cache, a processor, and a memory bridge shown in FIG. 1 during one example of a write flush command.
- FIG. 3 is a block diagram of one embodiment of a cache, a processor, and a memory bridge shown in FIG. 1 during another example of a write flush command.
- FIG. 4 is a flowchart illustrating operation of one embodiment of the cache shown in FIGS. 1 - 3 .
- FIG. 5 is a table illustrating an exemplary set of coherency commands and a table illustrating an exemplary set of transactions according to one embodiment of the node shown in FIG. 1.
- FIG. 6 is a block diagram of an address space supported by one embodiment of the node shown in FIG. 1.
- FIG. 7 is a decision tree illustrating operation of one embodiment of a node for a read transaction on the interconnect within the node.
- FIG. 8 is a decision tree illustrating operation of one embodiment of a node for a write transaction on the interconnect within the node.
- FIG. 9 is a diagram illustrating operation of one embodiment of the memory bridge for remote coherency commands received by the memory bridge.
- FIG. 10 is a block diagram of one embodiment of a computer accessible medium.
- FIG. 1 a block diagram of one embodiment of a node 10 is shown.
- the node 10 includes one or more processors 12 A- 12 N, a memory controller 14 , a switch 18 , a set of interface circuits 20 A- 20 C, a memory bridge 32 , and an L 2 cache 36 .
- the memory bridge 32 includes a remote line directory 34 .
- the node 10 includes an interconnect 22 to which the processors 12 A- 12 N, the memory controller 14 , the L 2 cache 36 , the memory bridge 32 , and the remote line directory 34 are coupled.
- the node 10 is coupled, through the memory controller 14 , to a memory 24 .
- the interface circuits 20 A- 20 C each include a receive (Rx) circuit 26 A- 26 C and a transmit (Tx) circuit 28 A- 28 C.
- the node 10 is coupled to a set of interfaces 30 A- 30 C through respective interface circuits 20 A- 20 C.
- the interface circuits 20 A- 20 C are coupled to the switch 18 , which is further coupled to the memory bridge 32 .
- a configuration register 38 is also illustrated in FIG. 1, which stores a node number (Node #) for the node 10 .
- the configuration register 38 is coupled to the L 2 cache 36 , the memory controller 14 , the memory bridge 32 , and the interface circuits 20 A- 20 C in the embodiment of FIG. 1. Additionally, the processors 12 A- 12 N may be coupled to receive the node number from the configuration register 38 .
- the node 10 may support intranode coherency for transactions on the interconnect 22 . Additionally, the node 10 may support internode coherency with other nodes (e.g. a CC-NUMA coherency, in one embodiment). Generally, as used herein, a memory bridge includes circuitry designed to handle internode coherency functions within a node. Particularly, in one embodiment, if a transaction on the interconnect 22 (e.g. a transaction issued by the processors 12 A- 12 N) accesses a cache block that is remote to the node 10 (i.e.
- the memory bridge 32 may issue one or more coherency commands to the other nodes to obtain the ownership (and a copy of the cache block, in some cases). Similarly, if the transaction access a local cache block but one or more other nodes have a copy of the cache block, the memory bridge 32 may issue coherency commands to other nodes. Still further, the memory bridge 32 may receive coherency commands from other nodes, and may perform transactions on the interconnect 22 to effect the coherency commands.
- a node such as node 10 may have memory coupled thereto (e.g. memory 24 ).
- the node may be responsible for tracking the state, in other nodes, of each cache block from the memory in that node.
- a node is referred to as the “home node” for cache blocks from the memory assigned to that node.
- a node is referred to as a “remote node” for a cache block if the node is not the home node for that cache block.
- a cache block is referred to as a local cache block in the home node for that cache block and as a remote cache block in other nodes.
- a remote node may begin the coherency process by requesting a copy of a cache block from the home node of that cache block using a coherency command.
- the memory bridge 32 in the remote node may detect a transaction on the interconnect 22 that accesses the cache block and may detect that the remote node does not have sufficient ownership of the cache block to complete the transaction (e.g. it may not have a copy of the cache block at all, or may have a shared copy and may require exclusive ownership to complete the transaction).
- the memory bridge 32 in the remote node may generate and transmit the coherency command to the home node to obtain the copy or to obtain sufficient ownership.
- the memory bridge 32 in the home node may determine if any state changes in other nodes are to be performed to grant the requested ownership to the remote node, and may transmit coherency commands (e.g. probe commands) to effect the state changes.
- coherency commands e.g. probe commands
- the memory bridge 32 in each node receiving the probe commands may effect the state changes and respond to the probe commands.
- the memory bridge 32 in the home node may respond to the remote node (e.g. with a fill command including the cache block).
- the remote line directory 34 may be used in the home node to track the state of the local cache blocks in the remote nodes.
- the remote line directory 34 is updated each time a cache block is transmitted to a remote node, the remote node returns the cache block to the home node, or the cache block is invalidated via probes.
- the “state” of a cache block in a given node refers to an indication of the ownership that the given node has for the cache block according to the coherency protocol implemented by the nodes. Certain levels of ownership may permit no access, read-only access, or read-write access to the cache block. For example, in one embodiment, the modified, shared, and invalid states are supported in the internode coherency protocol.
- the node may read and write the cache block and the node is responsible for returning the block to the home node if evicted from the node.
- the node may read the cache block but not write the cache block without transmitting a coherency command to the home node to obtain modified state for the cache block.
- the node may not read or write the cache block (i.e. the node does not have a valid copy of the cache block).
- Other embodiments may use other coherency protocols (e.g.
- agents within the node may implement the MESI protocol for intranode coherency.
- the node may be viewed as having a state in the internode coherency and individual agents may have a state in the intranode coherency (consistent with the internode coherency state for the node containing the agent).
- Coherency commands are transmitted and received on one of the interfaces 30 A- 30 C by the corresponding interface circuit 20 A- 20 C.
- the interface circuits 20 A- 20 C receive coherency commands for transmission from the memory bridge 32 and transmit coherency commands received from the interfaces 30 A- 30 C to the memory bridge 32 for processing, if the coherency commands require processing in the node 10 .
- a coherency command may be received that is passing through the node 10 to another node, and does not require processing in the node 10 .
- the interface circuits 20 A- 20 C may be configured to detect such commands and retransmit them (through another interface circuit 20 A- 20 C) without involving the memory bridge 32 .
- the interface circuits 20 A- 20 C are coupled to the memory bridge 32 through the switch 18 (although in other embodiments, the interface circuits 20 A- 20 C may have direct paths to the memory bridge 32 ).
- the switch 18 may selectively couple the interface circuits 20 A- 20 C (and particularly the Rx circuits 26 A- 26 C in the illustrated embodiment) to other interface circuits 20 A- 20 C (and particularly the Tx circuits 28 A- 28 C in the illustrated embodiment) or to the memory bridge 32 to transfer received coherency commands.
- the switch 18 may also selectively couple the memory bridge 32 to the interface circuits 20 A- 20 C (and particularly to the Tx circuits 28 A- 28 C in the illustrated embodiment) to transfer coherency commands generated by the memory bridge 32 from the memory bridge 32 to the interface circuits 20 A- 20 C for transmission on the corresponding interface 30 A- 30 C.
- the switch 18 may have request/grant interfaces to each of the interface circuits 20 A- 20 C and the memory bridge 32 for requesting transfers and granting those transfers.
- the switch 18 may have an input path from each source (the Rx circuits 26 A- 26 C and the memory bridge 32 ) and an output path to each destination (the Tx circuits 28 A- 28 C and the memory bridge 32 ), and may couple a granted input path to a granted output path for transmission of a coherency command (or a portion thereof, if coherency commands are larger than one transfer through the switch 18 ). The couplings may then be changed to the next granted input path and granted output path. Multiple independent input path/output path grants may occur concurrently.
- the interfaces 30 A- 30 C may support a set of virtual channels in which commands are transmitted. Each virtual channel is defined to flow independent of the other virtual channels, even though the virtual channels may share certain physical resources (e.g. the interface 30 A- 30 C on which the commands are flowing). These virtual channels may be mapped to internal virtual channels (referred to as switch virtual channels herein).
- the switch 18 may be virtual-channel aware. That is, the switch 18 may grant a coupling between a source and a destination based not only on the ability of the source to transfer data and the destination to receive data, but also on the ability of the source to transfer data in a particular switch virtual channel and the destination to receive data on that switch virtual channel. Thus, requests from sources may indicate the destination and the virtual channel on which data is to be transferred, and requests from destinations may indicate the virtual channel on which data may be received.
- a node may include one or more coherent agents (dotted enclosure 16 in FIG. 1).
- the processors 12 A- 12 N, the L 2 cache 36 , and the memory controller 14 may be examples of coherent agents 16 .
- the memory bridge 32 may be a coherent agent (on behalf of other nodes).
- other embodiments may include other coherent agents as well, such as a bridge to one or more I/O interface circuits, or the I/O interface circuits themselves.
- an agent includes any circuit which participates in transactions on an interconnect.
- a coherent agent is an agent that is capable of performing coherent transactions and operating in a coherent fashion with regard to transactions.
- a transaction is a communication on an interconnect.
- the transaction is sourced by one agent on the interconnect, and may have one or more agents as a target of the transaction.
- Read transactions specify a transfer of data from a target to the source, while write transactions specify a transfer of data from the source to the target.
- Other transactions may be used to communicate between agents without transfer of data, in some embodiments.
- Each of the interface circuits 20 A- 20 C are configured to receive and transmit on the respective interfaces 30 A- 30 C to which they are connected.
- the Rx circuits 26 A- 26 C handle the receiving of communications from the interfaces 30 A- 30 C
- the Tx circuits 28 A- 28 C handle the transmitting of communications on the interfaces 30 A- 30 C.
- Each of the interfaces 30 A- 30 C used for coherent communications are defined to be capable of transmitting and receiving coherency commands. Particularly, in the embodiment of FIG. 1, those interfaces 30 A- 30 C may be defined to receive/transmit coherency commands to and from the node 10 from other nodes. Additionally, other types of commands may be carried.
- each interface 30 A- 30 C may be a HyperTransportTM (HT) interface, including an extension to the HT interface to include coherency commands (HTcc). Additionally, in some embodiments, an extension to the HyperTransport interface to carry packet data (Packet over HyperTransport, or PoHT) may be supported.
- coherency commands include any communications between nodes that are used to maintain coherency between nodes.
- the commands may include read or write requests initiated by a node to fetch or update a cache block belonging to another node, probes to invalidate cached copies of cache blocks in remote nodes (and possibly to return a modified copy of the cache block to the home node), responses to probe commands, fills which transfer data, etc.
- one or more of the interface circuits 20 A- 20 C may not be used for coherency management and may be defined as packet interfaces.
- Such interfaces 30 A- 30 C may be HT interfaces.
- Such interfaces 30 A- 30 C may be system packet interfaces (SPI) according to any level of the SPI specification set forth by the Optical Internetworking Forum (e.g. level 3 , level 4 , or level 5 ).
- the interfaces may be SPI- 4 phase 2 interfaces.
- each interface circuit 20 A- 20 C may be configurable to communicate on either the SPI-4 interface or the HT interface.
- Each interface circuit 20 A- 20 C may be individually programmable, permitting various combinations of the HT and SPI-4 interfaces as interfaces 30 A- 30 C.
- the programming may be performed in any fashion (e.g. sampling certain signals during reset, shifting values into configuration registers (not shown) during reset, programming the interfaces with configuration space commands after reset, pins that are tied up or down externally to indicate the desired programming, etc.).
- Other embodiments may employ any interface capable of carrying packet data (e.g. the Media Independent Interface (MII) or the Gigabit MII (GMII) interfaces, X.25, Frame Relay, Asynchronous Transfer Mode (ATM), etc.).
- the packet interfaces may carry packet data directly (e.g. transmitting the packet data with various control information indicating the start of packet, end of packet, etc.) or indirectly (e.g. transmitting the packet data as a payload of a command, such as PoHT).
- the node 10 may also include a packet direct memory access (DMA) circuit configured to transfer packets to and from the memory 24 on behalf of the interface circuits 20 A- 20 C.
- the switch 18 may be used to transmit packet data from the interface circuits 20 A- 20 C to the packet DMA circuit and from the packet DMA circuit to the interface circuits 20 A- 20 C. Additionally, packets may be routed from an Rx circuit 26 A- 26 C to a Tx circuit 28 A- 28 C through the switch 18 , in some embodiments.
- DMA packet direct memory access
- the processors 12 A- 12 N may be designed to any instruction set architecture, and may execute programs written to that instruction set architecture.
- Exemplary instruction set architectures may include the MIPS instruction set architecture (including the MIPS-3D and MIPS MDMX application specific extensions), the IA-32 or IA-64 instruction set architectures developed by Intel Corp., the PowerPC instruction set architecture, the Alpha instruction set architecture, the ARM instruction set architecture, or any other instruction set architecture.
- the node 10 may include any number of processors (e.g. as few as one processor, two processors, four processors, etc.).
- the L 2 cache 36 may be any type and capacity of cache memory, employing any organization (e.g. set associative, direct mapped, fully associative, etc.). In one embodiment, the L 2 cache 36 may be an 8 way, set associative, 1 MB cache. The L 2 cache 36 is referred to as L 2 herein because the processors 12 A- 12 N may include internal (L 1 ) caches. In other embodiments the L 2 cache 36 may be an L 1 cache, an L 3 cache, or any other level as desired.
- the memory controller 14 is configured to access the memory 24 in response to read and write transactions received on the interconnect 22 .
- the memory controller 14 may receive a hit signal from the L 2 cache, and if a hit is detected in the L 2 cache for a given read/write transaction, the memory controller 14 may not respond to that transaction.
- the memory controller 14 may be designed to access any of a variety of types of memory.
- the memory controller 14 may be designed for synchronous dynamic random access memory (SDRAM), and more particularly double data rate (DDR) SDRAM.
- SDRAM synchronous dynamic random access memory
- DDR double data rate
- the memory controller 16 may be designed for DRAM, DDR synchronous graphics RAM (SGRAM), DDR fast cycle RAM (FCRAM), DDR-II SDRAM, Rambus DRAM (RDRAM), SRAM, or any other suitable memory device or combinations of the above mentioned memory devices.
- SGRAM DDR synchronous graphics RAM
- FCRAM DDR fast cycle RAM
- RDRAM Rambus DRAM
- SRAM SRAM
- the interconnect 22 may be any form of communication medium between the devices coupled to the interconnect.
- the interconnect 22 may include shared buses, crossbar connections, point-to-point connections in a ring, star, or any other topology, meshes, cubes, etc.
- the interconnect 22 may also include storage, in some embodiments.
- the interconnect 22 may comprise a bus.
- the bus may be a split transaction bus, in one embodiment (i.e. having separate address and data phases). The data phases of various transactions on the bus may proceed out of order with the address phases.
- the bus may also support coherency and thus may include a response phase to transmit coherency response information.
- the bus may employ a distributed arbitration scheme, in one embodiment.
- the bus may be pipelined.
- the bus may employ any suitable signaling technique.
- differential signaling may be used for high speed signal transmission.
- Other embodiments may employ any other signaling technique (e.g. TTL, CMOS, GTL, HSTL, etc.).
- Other embodiments may employ non-split transaction buses arbitrated with a single arbitration for address and data and/or a split transaction bus in which the data bus is not explicitly arbitrated. Either a central arbitration scheme or a distributed arbitration scheme may be used, according to design choice.
- the bus may not be pipelined, if desired.
- the node 10 may include additional circuitry, not shown in FIG. 1.
- the node 10 may include various I/O devices and/or interfaces.
- Exemplary I/O may include one or more PCI interfaces, one or more serial interfaces, Personal Computer Memory Card International Association (PCMCIA) interfaces, etc.
- PCMCIA Personal Computer Memory Card International Association
- Such interfaces may be directly coupled to the interconnect 22 or may be coupled through one or more I/O bridge circuits.
- the node 10 (and more particularly the processors 12 A- 12 N, the memory controller 14 , the L 2 cache 36 , the interface circuits 20 A- 20 C, the memory bridge 32 including the remote line directory 34 , the switch 18 , the configuration register 38 , and the interconnect 22 ) may be integrated onto a single integrated circuit as a system on a chip configuration.
- the additional circuitry mentioned above may also be integrated.
- other embodiments may implement one or more of the devices as separate integrated circuits.
- the memory 24 may be integrated as well.
- one or more of the components may be implemented as separate integrated circuits, or all components may be separate integrated circuits, as desired. Any level of integration may be used.
- an interface circuit includes any circuitry configured to communicate on an interface according to the protocol defined for the interface.
- the interface circuit may include receive circuitry configured to receive communications on the interface and transmit the received communications to other circuitry internal to the system that includes the interface circuit.
- the interface circuit may also include transmit circuitry configured to receive communications from the other circuitry internal to the system and configured to transmit the communications on the interface.
- cache blocks and maintaining coherency on a cache block granularity that is, each cache block has a coherency state that applies to the entire cache block as a unit.
- Other embodiments may maintain coherency on a different granularity than a cache block, which may be referred to as a coherency block.
- a coherency block may be smaller than a cache line, a cache line, or larger than a cache line, as desired.
- the discussion herein of cache blocks and maintaining coherency therefor applies equally to coherency blocks of any size.
- the node 10 may cache one or more remote cache blocks, and may have a state in the node 10 for the cache block (the “node state”).
- the node state may be the state in which the remote cache block is provided to the node 10 (e.g. shared or modified) that is tracked by the home node of the remote cache block. If the node state in the node 10 is modified, the node 10 returns the cache block to the home node when the node 10 evicts the cache block. In such cases, the home node may update the state for the remote cache block to indicate that the node 10 no longer has a copy. Accordingly, when evicting a remote cache block with a modified node state from the node 10 , the node 10 ensures that any copies of the remote cache block in the node 10 are invalidated.
- the node 10 may include a transaction for transferring an evicted remote cache block to the memory bridge 32 for return to its home node.
- the transaction may be defined to transfer the cache block addressed by the transaction to the agent responsible for internode coherency (e.g. the memory bridge 32 , in the embodiment of FIG. 1).
- the transaction (referred to as a write flush transaction, or WrFlush transaction, herein for convenience) may be used for evicting remote cache blocks, while other transactions (transactions with different command encodings on the interconnect 22 ) may be used to coherently or non-coherently transfer data among the agents (including the memory bridge 32 , for remote transactions).
- the WrFlush transaction may be received by other coherent agents within the node 10 , and the coherent agents may invalidate the remote cache block if stored therein in response to the WrFlush transaction. Additionally, if a coherent agent that did not initiate the WrFlush transaction has a modified copy of the remote cache block, that coherent agent may transmit the cache block during the data portion of the WrFlush transaction instead of the initiating agent. The coherent agent may signal the initiating agent (e.g. during a response portion of the transaction) that the coherent agent is to transfer the data.
- the node 10 There may be various coherent agents in the node 10 (e.g. the L 2 cache 36 , the processors 12 A- 12 N, etc.), one or more of which may have a copy of a given remote cache block. Multiple copies may exist when various agents share the remote cache block according to the intranode coherency scheme employed on the interconnect 22 .
- a MESI protocol may be employed, and may be enforced by the coherent agents snooping transactions on the interconnect 22 .
- Other embodiments may employ a MOESI coherency protocol or any other protocol.
- the ownership of the remote cache block may be not be determined by the initiating agent prior to initiating the WrFlush transaction.
- the WrFlush transaction permits an exclusive owner (according to the intranode coherent protocol) to transfer the cache block, or the initiating agent transfers the cache block if no other agent is the exclusive owner.
- the node 10 may include an agent designated for retaining the node state (that is, the state that the home node is tracking for the node 10 ) of a remote cache block.
- the L 2 cache 36 may be the designated agent (although any agent may be designated in other embodiments).
- the L 2 cache 36 may allocate a storage location to store the remote cache block. The state of the cache block is stored in the L 2 cache 36 as well.
- the L 2 cache 36 may permit other coherent agents to access the remote cache block and to cache the remote cache block in any state that is consistent with the node state (e.g. any state that is less permissive than the node state or has the same permissiveness as the node state, where read access is less permissive than write access or read/write access). For example, if the node state is modified, any state may be maintained by a caching node. If the node state is shared, only the shared or invalid states are consistent with the node state. Since other coherent agents may have a copy of a remote cache block, if the L 2 cache 36 evicts the remote cache block, the other coherent agents may be forced to evict the remote cache block as well.
- the node state e.g. any state that is less permissive than the node state or has the same permissiveness as the node state, where read access is less permissive than write access or read/write access.
- any state may be maintained by a caching node. If the
- the L 2 cache 36 may force the eviction if the node state is modified, but not if the node state is shared. However, since the L 2 cache 36 allows other coherent agents to cache the remote cache block in the modified state (and thus permits these coherent agents to modify the remote cache block), another coherent agent may have a copy of the remote cache block which is more up to date than the L 2 cache's copy.
- the L 2 cache 36 may use the WrFlush transaction to evict remote cache blocks having a modified node state.
- the L 2 cache 36 may initiate the WrFlush transaction on the interconnect 22 .
- the coherent agents may snoop the WrFlush transaction, and may invalidate any copies of the remote cache block in the coherent agents in response.
- Each coherent agent may signal a snoop response to the L 2 cache 36 during a response portion of the WrFlush transaction. If a coherent agent has a modified copy of the cache block, that coherent agent may signal exclusive in the response portion. That coherent agent may then become responsible for transmitting the remote cache block in the data portion of the WrFlush transaction.
- the L 2 cache 36 transmits the remote cache block in the data portion of the WrFlush transaction. In either case, the memory bridge 32 receives the WrFlush transaction and the remote cache block for transfer to the home node of the remote cache block.
- the L 2 cache 36 may use the WrFlush transaction to evict remote cache blocks to the memory bridge 32 and may use a different transaction (e.g. a write (Wr) transaction) to evict local cache blocks.
- the memory controller 14 may receive the Wr transactions and may update the memory 24 with the local cache blocks received from the L 2 cache 36 .
- the coherent agents do not snoop the Wr transaction.
- the memory controller 14 may not process the WrFlush transaction. For example, the memory controller 14 may detect the WrFlush transaction and ignore the transaction, or may detect that the transaction address a remote cache block and ignore the transaction.
- a coherent agent may not necessarily determine if it has a modified copy of the cache block to signal exclusive in the response portion.
- the processors 12 A- 12 N may maintain a set of snoop tags for the data caches, used for snooping purposes.
- the snoop tags may track the state of the tags in the data cache, except for the modified state. That is, the snoop tags may have invalid, shared, or exclusive states (where exclusive or modified in the data cache is represented by the exclusive state).
- a processors 12 A- 12 N may transfer the cache block for the WrFlush transaction if the processor has the cache block in either the exclusive or the modified state.
- the WrFlush transaction may be decoded by the snooping agents (e.g. the processors 12 A- 12 N) as an exclusive read transaction.
- the WrFlush transaction may have the same effect on the snooping agent as an exclusive read transaction would have.
- the WrFlush transaction may have other uses as well.
- the WrFlush transaction may be initiated by an initiating agent to transfer a cache block to a second agent, where the cache block may be transmitted by either the initiating agent or a third agent dependent upon the state of the cache block in the third agent.
- FIGS. 2 and 3 are block diagrams illustrating the L 2 cache 36 , the processor 12 A, and the memory bridge 32 for example operation of the WrFlush transaction according to one embodiment of the node 10 .
- FIG. 2 is an example of the WrFlush transaction in which the processor 12 A has the remote cache block in the shared state
- FIG. 3 is an example of the WrFlush transaction in which the processor 12 A has the remote cache block in the modified state.
- the remote cache block is addressed by the address “A 1 ” in the examples.
- the L 2 cache 36 has the remote cache block addressed by the address A 1 in the modified state. In other words, the node state of the remote cache block is modified. Additionally, the processor 12 A has a copy of the remote cache block in the shared state. The L 2 cache 36 evicts the cache block and, detecting that the cache block is a remote cache block, initiates a WrFlush transaction to transfer the remote cache block to the memory bridge 32 (arrow 40 ). Additionally, the processor 12 A snoops the WrFlush transaction, and detects that the remote cache block is cached in the shared state. The processor 12 A invalidates the remote cache block in its cache.
- the processor 12 A transmits a shared response to the L 2 cache 36 (arrow 42 ). Since there is not an exclusive response, the L 2 cache 36 transmits the remote cache block during the data portion of the transaction (arrow 44 ). The memory bridge 32 receives the remote cache block, and subsequently transmits the remote cache block to the home node (not shown in FIG. 2).
- the L 2 cache 36 has the remote cache block addressed by the address A 1 in the modified state.
- the node state of the remote cache block is modified.
- the processor 12 A has a copy of the remote cache block in the modified state.
- the processor 12 A's copy is more up to date than the L 2 cache's copy, and thus should be transmitted to the memory bridge 32 for return to the home node.
- the L 2 cache 36 evicts the cache block and, detecting that the cache block is a remote cache block, initiates a WrFlush transaction to transfer the remote cache block to the memory bridge 32 (arrow 50 ).
- the processor 12 A snoops the WrFlush transaction, and detects that the remote cache block is cached in the modified state.
- the processor 12 A invalidates the remote cache block in its cache.
- the processor 12 A transmits an exclusive response to the L 2 cache 36 (arrow 52 ). Since there is an exclusive response, the L 2 cache 36 inhibits transmission of the remote cache block during the data portion of the transaction. Instead, the processor 12 A transmits the remote cache block during the data portion (arrow 54 ).
- the memory bridge 32 receives the remote cache block, and subsequently transmits the remote cache block to the home node (not shown in FIG. 3).
- any coherent agent may transmit the remote cache block in the data portion of the WrFlush transaction (similar to FIG. 3). If no coherent agent responds exclusive, the data portion of the WrFlush transaction may proceed as in FIG. 2.
- the processors 12 A- 12 N and/or other coherent snooping agents may not determine whether or not a remote cache block is exclusive or modified to respond in the response phase. In such embodiments, the snooping agent may respond exclusive if the remote cache block is in the exclusive state or the modified state in the snooping agent, and may signal with the data transfer whether or not the remote cache block is modified.
- a given coherent agent may invalidate a cache block in its cache at a delayed time with regard to the snoop that causes the block to be invalidated.
- the snoop may be queued for later invalidation, and the queue may be checked in parallel with a cache access to ensure the invalidated cache block is not used.
- a “snoop” may include receiving a transaction initiated on the interconnect 22 , and checking for the state of the cache block addressed by the transaction in response to the transaction. The result of the snoop may be signaled (e.g. during a response portion of the transaction), and a state change may be effected in the snooping agent for the cache block addressed by the transaction (if appropriate based on the transaction and the current state, according to the coherency protocol). A copy of a cache block may be modified if the copy has been changed by the caching agent from the copy that was supplied to that caching agent.
- the L 2 cache 36 may be programmable to reserve one or more locations for storing remote cache blocks. Such embodiments may inhibit selecting the reserved locations for replacement to store local cache blocks that miss in the L 2 cache 36 .
- set associative embodiments may be programmable to reserve one or more ways for remote cache blocks.
- portions of the WrFlush transaction are referred to above (e.g. the response portion and the data portion).
- the portions may occur in various fashions.
- a packet-based interconnect may include a request packet to initiate the transaction (including the address A 1 ), one or more response packets indicating the response (which may be included in the response portion) and a data packet (which may be included in the data portion).
- the response portion and data portion may be phases on the bus (e.g. a split transaction bus may include an address phase on the address bus, a response phase on the response lines, and a data phase on the data bus).
- FIG. 4 a flowchart is shown illustrating operation of one embodiment of the L 2 cache 36 in response to determining that a cache block is to be evicted (e.g. in response to an L 2 cache miss for a different cache block).
- the blocks in FIG. 4 are illustrated in a particular order for ease of understanding, but any order may be used. Furthermore, blocks may be performed in parallel by combinatorial logic circuitry in the L 2 cache 36 . Still further, blocks may be pipelined over multiple clock cycles and/or the flowchart of FIG. 4 may represent operation over a multiple clock cycles.
- the L 2 cache 36 may determine if the evicted cache block is in the modified state (decision block 60 ). If the evicted cache block is not modified, in this embodiment, the L 2 cache 36 need not perform a transaction on the interconnect 22 irrespective of whether the evicted cache block is local or remote. Thus, if the evicted cache block is not in the modified state (decision block 60 —“no” leg), the L 2 cache 36 may drop the evicted cache block (block 62 ).
- the L 2 cache block may determine if the cache block is a remote cache block (decision block 64 ).
- a remote cache block may be detectable by its address. For example, one embodiment described below with regard to FIGS. 5 - 9 maps addresses to nodes based on the most significant bits of address and the node number in the node. Other embodiments may detect remote cache blocks in other fashions (e.g. the L 2 cache 36 may store an indication for each cache block, indicating if it is local or remote).
- the L 2 cache 36 initiates a write transaction (Wr transaction) on the interconnect 22 to transfer the local cache block to the memory controller 14 (block 66 ). This transaction may not be snooped, and thus the L 2 cache 36 may transmit the cache block during the data portion of the Wr transaction.
- Wr transaction write transaction
- the L 2 cache 36 initiates a WrFlush transaction on the interconnect 22 (block 68 ).
- the L 2 cache 36 may wait for the response portion of the WrFlush transaction and, if an exclusive response is received (decision block 70 —“yes” leg), the L 2 cache 36 may inhibit transmitting the cache block during the data portion.
- the L 2 cache 36 transmits the cache block to the memory bridge 32 during the data portion of the transaction (block 72 ).
- FIGS. 5 - 9 illustrate additional details regarding one exemplary embodiment of a CC-NUMA protocol that may be employed by one embodiment of the node 10 .
- the embodiment of FIGS. 5 - 9 is merely exemplary. Numerous other implementations of CC-NUMA protocols or other distributed memory system protocols may be used in other embodiments.
- a table 142 is shown illustrating an exemplary set of transactions supported by one embodiment of the interconnect 22 and a table 144 is shown illustrating an exemplary set of coherency commands supported by one embodiment of the interfaces 30 .
- Other embodiments including subsets, supersets, or alternative sets of commands may be used.
- An agent in the node 10 may read a cache block (either remote or local) using the read shared (RdShd) or read exclusive (RdExc) transactions on the interconnect 22 .
- the RdShd transaction is used to request a shared copy of the cache block
- the RdExc transaction is used to request an exclusive copy of the cache block. If the RdShd transaction is used, and no other agent reports having a copy of the cache block during the response phase of the transaction (except for the L 2 cache 36 and/or the memory controller 14 ), the agent may take the cache block in the exclusive state.
- other agents in the node invalidate their copies of the cache block (if any).
- an exclusive (or modified) owner of the cache block may supply the data for the transaction in the data phase.
- Other embodiments may employ other mechanisms (e.g. a retry on the interconnect 22 ) to ensure the transfer of a modified cache block.
- the write transaction (Wr) and the write invalidate transaction (WrInv) may be used by an agent to write a cache block to memory.
- the Wr transaction may be used by an owner having the modified state for the block, since no other copies of the block need to be invalidated.
- the WrInv transaction may be used by an agent that does not have exclusive ownership of the block (the agent may even have the invalid state for the block).
- the WrInv transaction causes other agents to invalidate any copies of the block, including modified copies.
- the WrInv transaction may be used by an agent that is writing the entire cache block. For example, a DMA that is writing the entire cache block with new data may use the transaction to avoid a read transaction followed by a write transaction.
- the RdKill and RdInv transactions may be used by the memory bridge 32 in response to probes received by the node 10 from other nodes.
- the RdKill and RdInv transactions cause the initiator (the memory bridge 32 ) to acquire exclusive access to the cache block and cause any cache agents to invalidate their copies (transferring data to the initiator similar to the RdShd and RdExc transactions).
- the RdKill transaction also cancels a reservation established by the load-linked instruction in the MIPS instruction set, while the RdInv transaction does not.
- a single transaction may be used for probes.
- the WrFlush transaction is a write transaction which may be initiated by an agent and another agent may have an exclusive or modified copy of the block.
- the other agent provides the data for the WrFlush transaction, or the initiating agent provides the data if no other agent has an exclusive or modified copy of the block.
- the WrFlush transaction may be used, in one embodiment as described above by the L 2 cache 36 .
- the Nop transaction is a no-operation transaction.
- the Nop may be used if an agent is granted use of the interconnect 22 (e.g. the address bus, in embodiments in which the interconnect 22 is a split transaction bus) and the agent determines that it no longer has a transaction to run on the interconnect 22 .
- the commands illustrated in the table 144 will next be described.
- the virtual channels may include, in the illustrated embodiment: the coherent read (CRd) virtual channel; the probe (Probe) virtual channel; the acknowledge (Ack) virtual channel; and coherent fill (CFill) virtual channel.
- the CRd, Probe, Ack, and CFill virtual channels are defined for the HTcc commands.
- There may be additional virtual channels for the standard HT commands e.g. non-posted command (NPC) virtual channel, the posted command (PC) virtual channel, and the response (RSP) virtual channel).
- NPC non-posted command
- PC posted command
- RSP response
- the cRdShd or cRdExc commands may be issued by the memory bridge 32 in response to a RdShd or RdExc transactions on the interconnect 22 , respectively, to read a remote cache block not stored in the node (or, in the case of RdExc, the block may be stored in the node but in the shared state). If the cache block is stored in the node (with exclusive ownership, in the case of the RdExc transaction), the read is completed on the interconnect 22 without any coherency command transmission by the memory bridge 32 .
- the Flush and Kill commands are probe commands for this embodiment.
- the memory bridge 32 at the home node of a cache block may issue probe commands in response to a cRdShd or cRdExc command.
- the memory bridge 32 at the home node of the cache block may also issue a probe command in response to a transaction for a local cache block, if one or more remote nodes has a copy of the cache block.
- the Flush command is used to request that a remote modified owner of a cache block return the cache block to the home node (and invalidate the cache block in the remote modified owner).
- the Kill command is used to request that a remote owner invalidate the cache block.
- additional probe commands may be supported for other state change requests (e.g. allowing remote owners to retain a shared copy of the cache block).
- the probe commands are responded to (after effecting the state changes requested by the probe commands) using either the Kill_Ack or WB commands.
- the Kill_Ack command is an acknowledgement that a Kill command has been processed by a receiving node.
- the WB command is a write back of the cache block, and is transmitted in response to the Flush command.
- the WB command may also be used by a node to write back a remote cache block that is being evicted from the node.
- the Fill command is the command to transfer data to a remote node that has transmitted a read command (cRdExc or cRdShd) to the home node.
- the Fill command is issued by the memory bridge 32 in the home node after the probes (if any) for a cache block have completed.
- FIG. 6 a block diagram illustrating one embodiment of an address space implemented by one embodiment of the node 10 is shown. Addresses shown in FIG. 6 are illustrated as hexadecimal digits, with an under bar (“_”) separating groups of four digits. Thus, in the embodiment illustrated in FIG. 6, 40 bits of address are supported. In other embodiments, more or fewer address bits may be supported.
- the address space between 00 — 0000 — 0000 and 0F_FFFF_FFFF is treated as local address space. Transactions generated by agents in the local address space do not generate coherency commands to other nodes, although coherency may be enforced within the node 10 for these addresses. That is, the local address space is not maintained coherent with other nodes. Various portions of the local address space may be memory mapped to I/O devices, HT, etc. as desired.
- the address space between 40 — 0000 — 0000 and EF_FFFF_FFFF is the remote coherent space 148 . That is, the address space between 40 — 0000 — 0000 and EF_FFFF_FFFF is maintained coherent between the nodes.
- Each node is assigned a portion of the remote coherent space, and that node is the home node for the portion. As shown in FIG. 1, each node is programmable with a node number. The node number is equal to the most significant nibble (4 bits) of the addresses for which that node is the home node, in this embodiment. Thus, the node numbers may range from 4 to E in the embodiment shown. Other embodiments may support more or fewer node numbers, as desired.
- each node is assigned a 64 Gigabyte (GB) portion of the memory space for which it is the home node.
- the size of the portion assigned to each node may be varied in other embodiments (e.g. based on the address size or other factors).
- the node having node number 5 aliases the address space 50 — 0000 — 0000 through 5F_FFFF_FFFF to 00 — 0000 — 0000 through 0F_FFFF_FFFF respectively (arrow 146 ).
- Internode coherent accesses to the memory 24 at the node 10 use the node-numbered address space (e.g. 50 — 0000 — 0000 to 5F_FFFF_FFFF, if the node number programmed into node 10 is 5 ) to access cache blocks in the memory 24 . That is agents in other nodes and agents within the node that are coherently accessing cache blocks in the memory use the remote coherent space, while access in the local address space are not maintained coherent with other nodes (even though the same cache block may be accessed). Thus the addresses are aliased, but not maintained coherent, in this embodiment. In other embodiments, the addresses in the remote coherent space and the corresponding addresses in the local address space may be maintained coherent.
- a cache block is referred to as local in a node if the cache block is part of the memory assigned to the node (as mentioned above).
- the cache block may be local if it is accessed from the local address space or the remote coherent space, as long as the address is in the range for which the node is the home node.
- a transaction on the interconnect 22 that accesses a local cache block may be referred to as a local transaction or local access.
- a transaction on the interconnect 22 that accesses a remote cache block via the remote coherent address space outside of the portion for which the node is the home node) may be referred to as a remote transaction or a remote access.
- the address space between 10 — 0000 — 0000 and 3F_FFFF_FFFF may be used for additional HT transactions (e.g. standard HT transactions) in the illustrated embodiment. Additionally, the address space between F0 — 0000 — 0000 and FF_FFFF_FFFF may be reserved in the illustrated embodiment.
- FIG. 7 a decision tree for a read transaction to a memory space address on the interconnect 22 of a node 10 is shown for one embodiment.
- the decision tree may illustrate operation of the node 10 for the read transaction for different conditions of the transaction, the state of the cache block accessed by the transaction, etc.
- the read transaction may, in one embodiment, include the RdShd, RdExc, RdKill, and RdInv transactions shown in the table 142 of FIG. 5.
- Each dot on the lines within the decision tree represents a divergence point of one or more limbs of the tree, which are labeled with the corresponding conditions. Where multiple limbs emerge from a dot, taking one limb also implies that the conditions for the other limbs are not met.
- the exclamation point (“!”) is used to indicate a logical NOT. Not shown in FIG. 7 is the state transition made by each coherent agent which is caching a copy of the cache block for the read transaction. If the read transaction is RdShd, the coherent agent may retain a copy of the cache block in the shared state. Otherwise, the coherent agent invalidates its copy of the cache block.
- the transaction may be either local or remote, as mentioned above. For local transactions, if the transaction is uncacheable, then a read from the memory 24 is performed (reference numeral 150 ). In one embodiment, the transaction may include an indication of whether or not the transaction is cacheable. If the transaction is uncacheable, it is treated as a non-coherent transaction in the present embodiment.
- each coherent agent responds with the state of the cache block in that agent.
- each coherent agent may have an associated shared (SHD) and exclusive (EXC) signal.
- the agent may signal invalid state by deasserting both the SHD and EXC signals.
- the agent may signal shared state by asserting the SHD signal and deasserting the EXC signal.
- the agent may signal exclusive state (or modified state) by asserting the EXC signal and deasserting the SHD signal.
- the exclusive and modified states may be treated the same in the response phase in this embodiment, and the exclusive/modified owner may provide the data.
- the exclusive/modified owner may provide, concurrent with the data, an indication of whether the state is exclusive or modified. While each agent may have its own SHD and EXC signals in this embodiment (and the initiating agent may receive the signals from each other agent), in other embodiments a shared SHD and EXC signal may be used by all agents.
- the memory controller may return a fatal error indication for the read transaction, in one embodiment. If the response is exclusive (SHD deasserted, EXC asserted) the exclusive owner provides the data for the read transaction on the interconnect 22 (reference numeral 154 ). If the exclusive owner is the memory bridge 32 (as recorded in the remote line directory 34 ), then a remote node has the cache block in the modified state. The memory bridge 32 issues a probe (Flush command) to retrieve the cache block from that remote node. The memory bridge 32 may supply the cache block returned from the remote node as the data for the read on the interconnect 22 .
- the response is shared (SHD asserted, EXC deasserted)
- the local transaction is RdExc
- the memory bridge 32 is one of the agents reporting shared
- at least one remote node may have a shared copy of the cache block.
- the memory bridge 32 may initiate a probe (Kill command) to invalidate the shared copies of the cache block in the remote node(s) (reference numeral 156 ).
- the data may be read from memory (or the L 2 cache 36 ) for this case, but the transfer of the data may be delayed until the remote node(s) have acknowledged the probe.
- the memory bridge 32 may signal the memory controller 14 /L 2 cache 36 when the acknowledgements have been received.
- each transaction may have a transaction identifier on the interconnect 22 .
- the memory bridge 32 may transmit the transaction identifier of the RdExc transaction to the memory controller 14 /L 2 cache 36 to indicate that the data may be transmitted.
- the L 2 cache 36 or the memory controller 14 may supply the data, depending on whether or not there is an L 2 hit for the cache block (reference numeral 158 ). Similarly, if the response is shared and the transaction is not RdExc, the L 2 cache 36 or the memory controller 14 may supply the data dependent on whether or not there is an L 2 hit for the cache block.
- the memory bridge 32 may generate a noncoherent read command on the interfaces 30 to read the data. For example, a standard HT read command may be used (reference numeral 160 ). If the remote transaction is cacheable and the response on the interconnect 22 is exclusive, then the exclusive owner supplies the data for the read (reference numeral 162 ). If the remote transaction is cacheable, the response is not exclusive, the cache block is an L 2 cache hit, and the transaction is either RdShd or the transaction is RdExc and the L 2 cache has the block in the modified state, then the L 2 cache 36 supplies the data for the read (reference numeral 164 ). Otherwise, the memory bridge 32 initiates a corresponding read command to the home node of the cache block (reference numeral 166 ).
- a standard HT read command may be used (reference numeral 160 ). If the remote transaction is cacheable and the response on the interconnect 22 is exclusive, then the exclusive owner supplies the data for the read (reference numeral 162 ). If the remote transaction is cacheable,
- FIG. 8 a decision tree for a write transaction to a memory space address on the interconnect 22 of a node 10 is shown for one embodiment.
- the decision tree may illustrate operation of the node for the write transaction for different conditions of the transaction, the state of the cache block accessed by the transaction, etc.
- the write transaction may, in one embodiment, include the Wr, WrInv, and WrFlush transactions shown in the table 142 of FIG. 5.
- Each dot on the lines within the decision tree represents a divergence point of one or more limbs of the tree, which are labeled with the corresponding conditions. Where multiple limbs emerge from a dot, taking one limb also implies that the conditions for the other limbs are not met.
- the exclamation point (“!”) is used to indicate a logical NOT.
- Not shown in FIG. 8 is the state transition made by each coherent agent which is caching a copy of the cache block for the write transaction. The coherent agent invalidates its copy of the cache block.
- the memory controller 14 (and the L 2 cache 36 , if an L 2 hit) updates with the write data (reference numeral 170 ). Additionally, the memory bridge 32 may generate probes to the remote nodes indicated by the remote line directory 34 . The update of the memory/L 2 cache may be delayed until the probes have been completed, at which time the memory bridge 32 may transmit the transaction identifier of the WrInv transaction to the L 2 cache 36 /memory controller 14 to permit the update.
- the memory controller 14 updates with the data (reference numeral 172 ). If the local transaction is cacheable, the memory controller 14 and/or the L 2 cache 36 updates with the data based on whether or not there is an L 2 cache hit (and, in some embodiments, based on an L 2 cache allocation indication in the transaction, which allows the source of the transaction to indicate whether or not the L 2 cache allocates a cache line for an L 2 cache miss) (reference numeral 174 ).
- the transaction is a remote transaction
- the transaction is a WrFlush transaction
- the response to the transaction is exclusive
- the exclusive owner supplies the data (reference numeral 176 ).
- the L 2 cache 36 supplies the data of the WrFlush transaction.
- the L 2 cache 36 retains the state of the node as recorded in the home node, and the L 2 cache 36 uses the WrFlush transaction to evict a remote cache block which is in the modified state in the node.
- another agent has the cache block in the exclusive state, that agent may have a more recent copy of the cache block that should be returned to the home node.
- the L 2 cache 36 supplies the block to be returned to the home node (reference numeral 182 ).
- the memory bridge 32 may capture the WrFlush transaction and data, and may perform a WB command to return the cache block to the home node.
- the memory bridge 32 receives the write transaction and performs a noncoherent Wr command (e.g. a standard HT write) to transmit the cache block to the home node (reference numeral 180 ). If the remote transaction is not a WrFlush transaction, is cache coherent, and is an L 2 hit, the L 2 cache 36 may update with the data (reference numeral 182 ).
- a noncoherent Wr command e.g. a standard HT write
- FIG. 9 a block diagram illustrating operation of one embodiment of the memory bridge 32 in response to various coherency commands received from the interface circuits 20 A- 20 C is shown.
- the received command is shown in an oval.
- Commands initiated by the memory bridge 32 in response to the received command are shown in solid boxes.
- Dotted boxes are commands received by the memory bridge 32 in response to the commands transmitted in the preceding solid boxes.
- the cache block affected by a command is shown in parentheses after the command.
- the remote line directory 34 may be accessed in response to a transaction on the interconnect 22 .
- the memory bridge 32 may initiate a transaction on the interconnect 22 in response to certain coherent commands in order to retrieve the remote line directory 34 (as well as to affect any state changes in the coherent agents coupled to the interconnect 22 , if applicable).
- the memory bridge 32 may be configured to read the remote line directory 34 prior to generating a transaction on the interconnect 22 , and may conditionally generate a transaction if needed based on the state of the remote line directory 34 for the requested cache block.
- the remote line directory 34 may maintain the remote state for a subset of the local cache blocks that are shareable remotely (e.g.
- a victim cache block may be replaced in the remote line directory 34 (and probes may be generated to invalidate the victim cache block in remote nodes).
- the remote line directory 34 may be configured to track the state of each cache block in the portion of the remote coherent space 148 that is assigned to the local node. In such embodiments, operations related to the victim cache blocks may be omitted from FIG. 9.
- the memory bridge 32 may generate a RdShd transaction on the interconnect 22 . Based on the remote line directory (RLD) state for the cache block A, a number of operations may occur. If the RLD state is shared, or invalid and there is an entry available for allocation without requiring a victim cache block to be evicted (“RLD empty” in FIG. 9), then the memory bridge 32 may transmit a fill command to the remote node with the data supplied to the memory bridge 32 in response to the RdShd transaction on the interconnect 22 (reference numeral 192 ).
- RLD remote line directory
- the memory bridge 32 may transmit probes to the remote nodes having copies of the victim cache block. If the victim cache block is shared, the memory bridge 32 may transmit a Kill command (or commands, if multiple nodes are sharing the victim cache block) for the victim block (reference numeral 194 ). The remote nodes respond with Kill_Ack commands for the victim block (reference numeral 196 ). If the victim block is modified, the memory bridge 32 may transmit a Flush command to the remote node having the modified state (reference numeral 198 ). The remote node may return the modified block with a WB command (reference numeral 200 ).
- a Kill command or commands, if multiple nodes are sharing the victim cache block
- the remote nodes respond with Kill_Ack commands for the victim block (reference numeral 196 ).
- the memory bridge 32 may transmit a Flush command to the remote node having the modified state (reference numeral 198 ).
- the remote node may return the modified block with a WB command (reference numeral 200 ).
- the memory bridge 32 may, in parallel, generate a Fill command for the cache block A (reference numeral 92 , via arrow 202 ). Finally, if the RLD state is modified for the cache block A, the memory bridge 32 may generate a Flush command for the cache block A to the remote node (reference numeral 204 ), which responds with a WB command and the cache block A (reference numeral 206 ). The memory bridge 32 may then transmit the Fill command with the cache block A provided via the write back command (reference numeral 192 ).
- operation may be similar to the cRdShd case for some RLD states. Similar to the cRdShd case, the memory bridge 32 may initiate a RdExc transaction on the interconnect 22 in response to the cRdExc command. Similar to the cRdShd case, if the RLD is invalid and no eviction of a victim cache block is needed in the RLD to allocate an entry for the cache block A, then the memory bridge 32 may supply the cache block supplied on the interconnect 22 for the RdExc transaction in a fill command to the remote node (reference numeral 212 ).
- the memory bridge 32 may operate in a similar fashion to the cRdShd case (reference numerals 214 and 216 and arrow 222 for the shared case of the victim block and reference numerals 218 and 220 and arrow 222 for the modified case of the victim block). If the RLD state is modified for the cache block A, the memory bridge 32 may operate in a similar fashion to the cRdShd case (reference numerals 224 and 226 ). If the RLD state is shared for the cache block A, the memory bridge 32 may generate Kill commands for each remote sharing node (reference numeral 228 ). The memory bridge 32 may wait for the Kill_Ack commands from the remote sharing nodes (reference numeral 230 ), and then transmit the Fill command with the cache block A provided on the interconnect 22 in response to the RdExc transaction (reference numeral 212 ).
- the memory bridge 32 may generate a Wr transaction on the interconnect 22 (reference numeral 240 ). If the RLD state is invalid for the cache block A, the memory bridge 32 may transmit the write data on the interconnect 22 and the Wr command is complete (reference numeral 242 ). If the RLD state is shared for the cache block A, the memory bridge 32 may generate Kill commands to each remote sharing node (reference numeral 244 ) and collect the Kill_Ack commands from those remote nodes (reference numeral 246 ) in addition to transmitting the data on the interconnect 22 .
- the memory bridge 32 may generate a Flush command to the remote node (reference numeral 248 ) and receive the WB command from the remote node (reference numeral 250 ). In one embodiment, the memory bridge 32 may delay transmitting the write data on the interconnect 22 until the WB command or Kill_Ack commands are received (although the data returned with the WB command may be dropped by the memory bridge 32 ).
- the above commands are received by the memory bridge 32 for cache blocks for which the node 10 including the memory bridge 32 is the home node.
- the memory bridge 32 may also receive Flush commands or Kill commands for cache blocks for which the node 10 is a remote node.
- the memory bridge 32 may initiate a RdInv transaction on the interconnect 22 . If the local state of the cache block is modified, the memory bridge 32 may transmit a WB command to the home node, with the cache block supplied on the interconnect 22 in response to the Rdlnv transaction (reference numeral 262 ).
- the memory bridge 32 may not respond to the Flush command (reference numeral 264 ). In this case, the node may already have transmitted a WB command to the home node (e.g. in response to evicting the cache block locally).
- the memory bridge 32 may initiate a RdKill transaction on the interconnect 22 .
- the memory bridge 32 may respond to the Kill command with a Kill_Ack command (reference numeral 272 ).
- the memory bridge 32 may also be configured to receive a non-cacheable read (RdNC) command (e.g. corresponding to a standard HT read) (reference numeral 280 ). In response, the memory bridge 32 may initiate a RdShd transaction on the interconnect 22 . If the RLD state is modified for the cache block including the data to be read, the memory bridge 32 may transmit a Flush command to the remote node having the modified cache block (reference numeral 282 ), and may receive the WB command from the remote node (reference numeral 284 ). Additionally, the memory bridge 32 may supply data received on the interconnect 22 in response to the RdShd transaction as a read response (RSP) to the requesting node (reference numeral 286 ).
- RdNC non-cacheable read
- RRP read response
- a block diagram of a computer accessible medium 300 including one or more data structures representative of the circuitry included in the node 10 is shown.
- a computer accessible medium may include storage media such as magnetic or optical media, e.g., disk, CD-ROM, or DVD-ROM, volatile or non-volatile memory media such as RAM (e.g. SDRAM, RDRAM, SRAM, etc.), ROM, etc., as well as media accessible via transmission media or signals such as electrical, electromagnetic, or digital signals, conveyed via a communication medium such as a network and/or a wireless link.
- the data structure(s) of the circuitry on the computer accessible medium 300 may be read by a program and used, directly or indirectly, to fabricate the hardware comprising the circuitry.
- the data structure(s) may include one or more behavioral-level descriptions or register-transfer level (RTL) descriptions of the hardware functionality in a high level design language (HDL) such as Verilog or VHDL.
- HDL high level design language
- the description(s) may be read by a synthesis tool which may synthesize the description to produce one or more netlist(s) comprising lists of gates from a synthesis library.
- the netlist(s) comprise a set of gates which also represent the functionality of the hardware comprising the circuitry.
- the netlist(s) may then be placed and routed to produce one or more data set(s) describing geometric shapes to be applied to masks.
- the masks may then be used in various semiconductor fabrication steps to produce a semiconductor circuit or circuits corresponding to the circuitry.
- the data structure(s) on computer accessible medium 300 may be the netlist(s) (with or without the synthesis library) or the data set(s), as desired.
- the data structures may comprise the output of a schematic program, or netlist(s) or data set(s) derived therefrom.
- computer accessible medium 300 includes a representation of the node 10
- other embodiments may include a representation of any portion of the node 10 (e.g. processors 12 A- 12 N, memory controller 14 , L 2 cache 36 , interconnect 22 , memory bridge 32 , remote line directory 34 , switch 18 , interface circuits 22 A- 22 C, etc.).
Abstract
Description
- This application claims benefit of priority to U.S. Provisional Patent Application Serial No. 60/380,740, filed May 15, 2002. This application is a continuation in part of U.S. patent application Ser. No. 10/270,028, filed on Oct. 11, 2002.
- 1. Field of the Invention
- This invention is related to coherent memory systems, including coherent distributed memory systems such as cache-coherent nonuniform memory access (CC-NUMA) memory systems.
- 2. Description of the Related Art
- Memory systems (including main memory and any caches in the system) are often designed to be coherent. That is, even though multiple copies of data from a given memory location may exist in the memory system, a read of that memory location returns the most recent data written to that memory location. The most recent data written to that memory location may be determined via an order of accesses to the memory location established according to the coherency mechanism. Typically, a coherent system may include one or more coherent agents and a memory controller coupled via an interconnect of some kind.
- One mechanism for scaling coherent systems to larger numbers of coherent agents is a distributed memory system. In such a system, memory is distributed among various nodes (which may also include coherent agents), and the nodes are interconnected. A coherent agent in one node may access memory in another node. One class of techniques for maintaining coherency in a distributed memory system is referred to as cache-coherent, nonuniform memory access (CC-NUMA). In a CC-NUMA system, access to memory may have a varying latency (e.g. memory in the same node as an agent may be accessed more rapidly than memory in another node, and accesses to different nodes may have varying latencies as well), but coherency is maintained. Data from another node may be cached in a given node.
- In some cases, it may be desirable to designate an agent in a node to maintain the state of data transferred to the node from another node. If that agent is to discard the data, a mechanism is desired to ensure that the data is deleted from other agents in the node. Additionally, the mechanism may be required to cause the most recently updated copy of the data, wherever it may reside within the node, to be returned to the node from which the data was read.
- In one embodiment, a node comprises a first agent, a second agent, and a third agent, all coupled to an interconnect. The first agent is configured to initiate a transaction on the interconnect to transfer a coherency block to the second agent. The third agent is configured to transmit the coherency block on the interconnect during a data portion of the transaction instead of the first agent responsive to a state of the coherency block in the third agent. In some embodiments, the first agent may be designated to store the node state of a remote cache block, and the second agent may be responsible for internode coherency within the node.
- In another embodiment, a node includes an interconnect and a cache coupled to the interconnect. The cache is configured to evict a cache block stored therein. The cache is configured to initiate a first transaction to transfer the cache block on the interconnect responsive to the cache block being a local cache block, and is configured to initiate a second transaction to transfer the cache block on the interconnect responsive to the cache block being a remote cache block. The second transaction is different from the first transaction.
- The following detailed description makes reference to the accompanying drawings, which are now briefly described.
- FIG. 1 is a block diagram of one embodiment of a node.
- FIG. 2 is a block diagram of one embodiment of a cache, a processor, and a memory bridge shown in FIG. 1 during one example of a write flush command.
- FIG. 3 is a block diagram of one embodiment of a cache, a processor, and a memory bridge shown in FIG. 1 during another example of a write flush command.
- FIG. 4 is a flowchart illustrating operation of one embodiment of the cache shown in FIGS.1-3.
- FIG. 5 is a table illustrating an exemplary set of coherency commands and a table illustrating an exemplary set of transactions according to one embodiment of the node shown in FIG. 1.
- FIG. 6 is a block diagram of an address space supported by one embodiment of the node shown in FIG. 1.
- FIG. 7 is a decision tree illustrating operation of one embodiment of a node for a read transaction on the interconnect within the node.
- FIG. 8 is a decision tree illustrating operation of one embodiment of a node for a write transaction on the interconnect within the node.
- FIG. 9 is a diagram illustrating operation of one embodiment of the memory bridge for remote coherency commands received by the memory bridge.
- FIG. 10 is a block diagram of one embodiment of a computer accessible medium.
- While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the drawings and detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the intention is to cover all modifications, equivalents and alternatives falling within the spirit and scope of the present invention as defined by the appended claims.
- Node Overview
- Turning now to FIG. 1, a block diagram of one embodiment of a
node 10 is shown. In the embodiment of FIG. 1, thenode 10 includes one ormore processors 12A-12N, amemory controller 14, aswitch 18, a set ofinterface circuits 20A-20C, amemory bridge 32, and anL2 cache 36. Thememory bridge 32 includes aremote line directory 34. Thenode 10 includes aninterconnect 22 to which theprocessors 12A-12N, thememory controller 14, theL2 cache 36, thememory bridge 32, and theremote line directory 34 are coupled. Thenode 10 is coupled, through thememory controller 14, to amemory 24. Theinterface circuits 20A-20C each include a receive (Rx)circuit 26A-26C and a transmit (Tx)circuit 28A-28C. Thenode 10 is coupled to a set ofinterfaces 30A-30C throughrespective interface circuits 20A-20C. Theinterface circuits 20A-20C are coupled to theswitch 18, which is further coupled to thememory bridge 32. A configuration register 38 is also illustrated in FIG. 1, which stores a node number (Node #) for thenode 10. The configuration register 38 is coupled to theL2 cache 36, thememory controller 14, thememory bridge 32, and theinterface circuits 20A-20C in the embodiment of FIG. 1. Additionally, theprocessors 12A-12N may be coupled to receive the node number from the configuration register 38. - The
node 10 may support intranode coherency for transactions on theinterconnect 22. Additionally, thenode 10 may support internode coherency with other nodes (e.g. a CC-NUMA coherency, in one embodiment). Generally, as used herein, a memory bridge includes circuitry designed to handle internode coherency functions within a node. Particularly, in one embodiment, if a transaction on the interconnect 22 (e.g. a transaction issued by theprocessors 12A-12N) accesses a cache block that is remote to the node 10 (i.e. the cache block is part of the memory coupled to a different node) and thenode 10 does not have sufficient ownership to perform the transaction, thememory bridge 32 may issue one or more coherency commands to the other nodes to obtain the ownership (and a copy of the cache block, in some cases). Similarly, if the transaction access a local cache block but one or more other nodes have a copy of the cache block, thememory bridge 32 may issue coherency commands to other nodes. Still further, thememory bridge 32 may receive coherency commands from other nodes, and may perform transactions on theinterconnect 22 to effect the coherency commands. - In one embodiment, a node such as
node 10 may have memory coupled thereto (e.g. memory 24). The node may be responsible for tracking the state, in other nodes, of each cache block from the memory in that node. A node is referred to as the “home node” for cache blocks from the memory assigned to that node. A node is referred to as a “remote node” for a cache block if the node is not the home node for that cache block. Similarly, a cache block is referred to as a local cache block in the home node for that cache block and as a remote cache block in other nodes. - Generally, a remote node may begin the coherency process by requesting a copy of a cache block from the home node of that cache block using a coherency command. The
memory bridge 32 in the remote node, for example, may detect a transaction on theinterconnect 22 that accesses the cache block and may detect that the remote node does not have sufficient ownership of the cache block to complete the transaction (e.g. it may not have a copy of the cache block at all, or may have a shared copy and may require exclusive ownership to complete the transaction). Thememory bridge 32 in the remote node may generate and transmit the coherency command to the home node to obtain the copy or to obtain sufficient ownership. Thememory bridge 32 in the home node may determine if any state changes in other nodes are to be performed to grant the requested ownership to the remote node, and may transmit coherency commands (e.g. probe commands) to effect the state changes. Thememory bridge 32 in each node receiving the probe commands may effect the state changes and respond to the probe commands. Once the responses have been received, thememory bridge 32 in the home node may respond to the remote node (e.g. with a fill command including the cache block). - The
remote line directory 34 may be used in the home node to track the state of the local cache blocks in the remote nodes. Theremote line directory 34 is updated each time a cache block is transmitted to a remote node, the remote node returns the cache block to the home node, or the cache block is invalidated via probes. As used herein, the “state” of a cache block in a given node refers to an indication of the ownership that the given node has for the cache block according to the coherency protocol implemented by the nodes. Certain levels of ownership may permit no access, read-only access, or read-write access to the cache block. For example, in one embodiment, the modified, shared, and invalid states are supported in the internode coherency protocol. In the modified state, the node may read and write the cache block and the node is responsible for returning the block to the home node if evicted from the node. In the shared state, the node may read the cache block but not write the cache block without transmitting a coherency command to the home node to obtain modified state for the cache block. In the invalid state, the node may not read or write the cache block (i.e. the node does not have a valid copy of the cache block). Other embodiments may use other coherency protocols (e.g. the MESI protocol, which includes the modified, shared, and invalid states and an exclusive state in which the cache block has not yet been updated but the node is permitted to read and write the cache block, or the MOESI protocol which includes the modified, exclusive, shared, and invalid states and an owned state which indicates that there may be shared copies of the block but the copy in main memory is stale). In one embodiment, agents within the node may implement the MESI protocol for intranode coherency. Thus, the node may be viewed as having a state in the internode coherency and individual agents may have a state in the intranode coherency (consistent with the internode coherency state for the node containing the agent). - Coherency commands are transmitted and received on one of the
interfaces 30A-30C by the correspondinginterface circuit 20A-20C. Theinterface circuits 20A-20C receive coherency commands for transmission from thememory bridge 32 and transmit coherency commands received from theinterfaces 30A-30C to thememory bridge 32 for processing, if the coherency commands require processing in thenode 10. In some embodiments, a coherency command may be received that is passing through thenode 10 to another node, and does not require processing in thenode 10. Theinterface circuits 20A-20C may be configured to detect such commands and retransmit them (through anotherinterface circuit 20A-20C) without involving thememory bridge 32. - In the illustrated embodiment, the
interface circuits 20A-20C are coupled to thememory bridge 32 through the switch 18 (although in other embodiments, theinterface circuits 20A-20C may have direct paths to the memory bridge 32). Theswitch 18 may selectively couple theinterface circuits 20A-20C (and particularly theRx circuits 26A-26C in the illustrated embodiment) toother interface circuits 20A-20C (and particularly theTx circuits 28A-28C in the illustrated embodiment) or to thememory bridge 32 to transfer received coherency commands. Theswitch 18 may also selectively couple thememory bridge 32 to theinterface circuits 20A-20C (and particularly to theTx circuits 28A-28C in the illustrated embodiment) to transfer coherency commands generated by thememory bridge 32 from thememory bridge 32 to theinterface circuits 20A-20C for transmission on thecorresponding interface 30A-30C. Theswitch 18 may have request/grant interfaces to each of theinterface circuits 20A-20C and thememory bridge 32 for requesting transfers and granting those transfers. Theswitch 18 may have an input path from each source (theRx circuits 26A-26C and the memory bridge 32) and an output path to each destination (theTx circuits 28A-28C and the memory bridge 32), and may couple a granted input path to a granted output path for transmission of a coherency command (or a portion thereof, if coherency commands are larger than one transfer through the switch 18). The couplings may then be changed to the next granted input path and granted output path. Multiple independent input path/output path grants may occur concurrently. - In one embodiment, the
interfaces 30A-30C may support a set of virtual channels in which commands are transmitted. Each virtual channel is defined to flow independent of the other virtual channels, even though the virtual channels may share certain physical resources (e.g. theinterface 30A-30C on which the commands are flowing). These virtual channels may be mapped to internal virtual channels (referred to as switch virtual channels herein). Theswitch 18 may be virtual-channel aware. That is, theswitch 18 may grant a coupling between a source and a destination based not only on the ability of the source to transfer data and the destination to receive data, but also on the ability of the source to transfer data in a particular switch virtual channel and the destination to receive data on that switch virtual channel. Thus, requests from sources may indicate the destination and the virtual channel on which data is to be transferred, and requests from destinations may indicate the virtual channel on which data may be received. - Generally speaking, a node may include one or more coherent agents (dotted
enclosure 16 in FIG. 1). In the embodiment of FIG. 1, theprocessors 12A-12N, theL2 cache 36, and thememory controller 14 may be examples ofcoherent agents 16. Additionally, thememory bridge 32 may be a coherent agent (on behalf of other nodes). However, other embodiments may include other coherent agents as well, such as a bridge to one or more I/O interface circuits, or the I/O interface circuits themselves. Generally, an agent includes any circuit which participates in transactions on an interconnect. A coherent agent is an agent that is capable of performing coherent transactions and operating in a coherent fashion with regard to transactions. A transaction is a communication on an interconnect. The transaction is sourced by one agent on the interconnect, and may have one or more agents as a target of the transaction. Read transactions specify a transfer of data from a target to the source, while write transactions specify a transfer of data from the source to the target. Other transactions may be used to communicate between agents without transfer of data, in some embodiments. - Each of the
interface circuits 20A-20C are configured to receive and transmit on therespective interfaces 30A-30C to which they are connected. TheRx circuits 26A-26C handle the receiving of communications from theinterfaces 30A-30C, and theTx circuits 28A-28C handle the transmitting of communications on theinterfaces 30A-30C. - Each of the
interfaces 30A-30C used for coherent communications are defined to be capable of transmitting and receiving coherency commands. Particularly, in the embodiment of FIG. 1, thoseinterfaces 30A-30C may be defined to receive/transmit coherency commands to and from thenode 10 from other nodes. Additionally, other types of commands may be carried. In one embodiment, eachinterface 30A-30C may be a HyperTransport™ (HT) interface, including an extension to the HT interface to include coherency commands (HTcc). Additionally, in some embodiments, an extension to the HyperTransport interface to carry packet data (Packet over HyperTransport, or PoHT) may be supported. As used herein, coherency commands include any communications between nodes that are used to maintain coherency between nodes. The commands may include read or write requests initiated by a node to fetch or update a cache block belonging to another node, probes to invalidate cached copies of cache blocks in remote nodes (and possibly to return a modified copy of the cache block to the home node), responses to probe commands, fills which transfer data, etc. - In some embodiments, one or more of the
interface circuits 20A-20C may not be used for coherency management and may be defined as packet interfaces.Such interfaces 30A-30C may be HT interfaces. Alternative,such interfaces 30A-30C may be system packet interfaces (SPI) according to any level of the SPI specification set forth by the Optical Internetworking Forum (e.g. level 3,level 4, or level 5). In one particular embodiment, the interfaces may be SPI-4 phase 2 interfaces. In the illustrated embodiment, eachinterface circuit 20A-20C may be configurable to communicate on either the SPI-4 interface or the HT interface. Eachinterface circuit 20A-20C may be individually programmable, permitting various combinations of the HT and SPI-4 interfaces asinterfaces 30A-30C. The programming may be performed in any fashion (e.g. sampling certain signals during reset, shifting values into configuration registers (not shown) during reset, programming the interfaces with configuration space commands after reset, pins that are tied up or down externally to indicate the desired programming, etc.). Other embodiments may employ any interface capable of carrying packet data (e.g. the Media Independent Interface (MII) or the Gigabit MII (GMII) interfaces, X.25, Frame Relay, Asynchronous Transfer Mode (ATM), etc.). The packet interfaces may carry packet data directly (e.g. transmitting the packet data with various control information indicating the start of packet, end of packet, etc.) or indirectly (e.g. transmitting the packet data as a payload of a command, such as PoHT). - In embodiments which also support packet traffic, the
node 10 may also include a packet direct memory access (DMA) circuit configured to transfer packets to and from thememory 24 on behalf of theinterface circuits 20A-20C. Theswitch 18 may be used to transmit packet data from theinterface circuits 20A-20C to the packet DMA circuit and from the packet DMA circuit to theinterface circuits 20A-20C. Additionally, packets may be routed from anRx circuit 26A-26C to aTx circuit 28A-28C through theswitch 18, in some embodiments. - The
processors 12A-12N may be designed to any instruction set architecture, and may execute programs written to that instruction set architecture. Exemplary instruction set architectures may include the MIPS instruction set architecture (including the MIPS-3D and MIPS MDMX application specific extensions), the IA-32 or IA-64 instruction set architectures developed by Intel Corp., the PowerPC instruction set architecture, the Alpha instruction set architecture, the ARM instruction set architecture, or any other instruction set architecture. Thenode 10 may include any number of processors (e.g. as few as one processor, two processors, four processors, etc.). - The
L2 cache 36 may be any type and capacity of cache memory, employing any organization (e.g. set associative, direct mapped, fully associative, etc.). In one embodiment, theL2 cache 36 may be an 8 way, set associative, 1 MB cache. TheL2 cache 36 is referred to as L2 herein because theprocessors 12A-12N may include internal (L1) caches. In other embodiments theL2 cache 36 may be an L1 cache, an L3 cache, or any other level as desired. - The
memory controller 14 is configured to access thememory 24 in response to read and write transactions received on theinterconnect 22. Thememory controller 14 may receive a hit signal from the L2 cache, and if a hit is detected in the L2 cache for a given read/write transaction, thememory controller 14 may not respond to that transaction. Thememory controller 14 may be designed to access any of a variety of types of memory. For example, thememory controller 14 may be designed for synchronous dynamic random access memory (SDRAM), and more particularly double data rate (DDR) SDRAM. Alternatively, thememory controller 16 may be designed for DRAM, DDR synchronous graphics RAM (SGRAM), DDR fast cycle RAM (FCRAM), DDR-II SDRAM, Rambus DRAM (RDRAM), SRAM, or any other suitable memory device or combinations of the above mentioned memory devices. - The
interconnect 22 may be any form of communication medium between the devices coupled to the interconnect. For example, in various embodiments, theinterconnect 22 may include shared buses, crossbar connections, point-to-point connections in a ring, star, or any other topology, meshes, cubes, etc. Theinterconnect 22 may also include storage, in some embodiments. In one particular embodiment, theinterconnect 22 may comprise a bus. The bus may be a split transaction bus, in one embodiment (i.e. having separate address and data phases). The data phases of various transactions on the bus may proceed out of order with the address phases. The bus may also support coherency and thus may include a response phase to transmit coherency response information. The bus may employ a distributed arbitration scheme, in one embodiment. In one embodiment, the bus may be pipelined. The bus may employ any suitable signaling technique. For example, in one embodiment, differential signaling may be used for high speed signal transmission. Other embodiments may employ any other signaling technique (e.g. TTL, CMOS, GTL, HSTL, etc.). Other embodiments may employ non-split transaction buses arbitrated with a single arbitration for address and data and/or a split transaction bus in which the data bus is not explicitly arbitrated. Either a central arbitration scheme or a distributed arbitration scheme may be used, according to design choice. Furthermore, the bus may not be pipelined, if desired. - Various embodiments of the
node 10 may include additional circuitry, not shown in FIG. 1. For example, thenode 10 may include various I/O devices and/or interfaces. Exemplary I/O may include one or more PCI interfaces, one or more serial interfaces, Personal Computer Memory Card International Association (PCMCIA) interfaces, etc. Such interfaces may be directly coupled to theinterconnect 22 or may be coupled through one or more I/O bridge circuits. - In one embodiment, the node10 (and more particularly the
processors 12A-12N, thememory controller 14, theL2 cache 36, theinterface circuits 20A-20C, thememory bridge 32 including theremote line directory 34, theswitch 18, the configuration register 38, and the interconnect 22) may be integrated onto a single integrated circuit as a system on a chip configuration. The additional circuitry mentioned above may also be integrated. Alternatively, other embodiments may implement one or more of the devices as separate integrated circuits. In another configuration, thememory 24 may be integrated as well. Alternatively, one or more of the components may be implemented as separate integrated circuits, or all components may be separate integrated circuits, as desired. Any level of integration may be used. - It is noted that, while three
interface circuits 20A-20C are illustrated in FIG. 1, one or more interface circuits may be implemented in various embodiments. As used herein, an interface circuit includes any circuitry configured to communicate on an interface according to the protocol defined for the interface. The interface circuit may include receive circuitry configured to receive communications on the interface and transmit the received communications to other circuitry internal to the system that includes the interface circuit. The interface circuit may also include transmit circuitry configured to receive communications from the other circuitry internal to the system and configured to transmit the communications on the interface. - It is noted that the discussion herein may describe cache blocks and maintaining coherency on a cache block granularity (that is, each cache block has a coherency state that applies to the entire cache block as a unit). Other embodiments may maintain coherency on a different granularity than a cache block, which may be referred to as a coherency block. A coherency block may be smaller than a cache line, a cache line, or larger than a cache line, as desired. The discussion herein of cache blocks and maintaining coherency therefor applies equally to coherency blocks of any size.
- Write Flush Transaction
- The
node 10 may cache one or more remote cache blocks, and may have a state in thenode 10 for the cache block (the “node state”). The node state may be the state in which the remote cache block is provided to the node 10 (e.g. shared or modified) that is tracked by the home node of the remote cache block. If the node state in thenode 10 is modified, thenode 10 returns the cache block to the home node when thenode 10 evicts the cache block. In such cases, the home node may update the state for the remote cache block to indicate that thenode 10 no longer has a copy. Accordingly, when evicting a remote cache block with a modified node state from thenode 10, thenode 10 ensures that any copies of the remote cache block in thenode 10 are invalidated. - The
node 10 may include a transaction for transferring an evicted remote cache block to thememory bridge 32 for return to its home node. The transaction may be defined to transfer the cache block addressed by the transaction to the agent responsible for internode coherency (e.g. thememory bridge 32, in the embodiment of FIG. 1). The transaction (referred to as a write flush transaction, or WrFlush transaction, herein for convenience) may be used for evicting remote cache blocks, while other transactions (transactions with different command encodings on the interconnect 22) may be used to coherently or non-coherently transfer data among the agents (including thememory bridge 32, for remote transactions). The WrFlush transaction may be received by other coherent agents within thenode 10, and the coherent agents may invalidate the remote cache block if stored therein in response to the WrFlush transaction. Additionally, if a coherent agent that did not initiate the WrFlush transaction has a modified copy of the remote cache block, that coherent agent may transmit the cache block during the data portion of the WrFlush transaction instead of the initiating agent. The coherent agent may signal the initiating agent (e.g. during a response portion of the transaction) that the coherent agent is to transfer the data. - There may be various coherent agents in the node10 (e.g. the
L2 cache 36, theprocessors 12A-12N, etc.), one or more of which may have a copy of a given remote cache block. Multiple copies may exist when various agents share the remote cache block according to the intranode coherency scheme employed on theinterconnect 22. For example, in one embodiment, a MESI protocol may be employed, and may be enforced by the coherent agents snooping transactions on theinterconnect 22. Other embodiments may employ a MOESI coherency protocol or any other protocol. Thus, when a remote cache block is being evicted from thenode 10, the ownership of the remote cache block may be not be determined by the initiating agent prior to initiating the WrFlush transaction. The WrFlush transaction permits an exclusive owner (according to the intranode coherent protocol) to transfer the cache block, or the initiating agent transfers the cache block if no other agent is the exclusive owner. - In one embodiment, the
node 10 may include an agent designated for retaining the node state (that is, the state that the home node is tracking for the node 10) of a remote cache block. For example, theL2 cache 36 may be the designated agent (although any agent may be designated in other embodiments). When thenode 10 fetches a remote cache block into thenode 10, theL2 cache 36 may allocate a storage location to store the remote cache block. The state of the cache block is stored in theL2 cache 36 as well. - The
L2 cache 36 may permit other coherent agents to access the remote cache block and to cache the remote cache block in any state that is consistent with the node state (e.g. any state that is less permissive than the node state or has the same permissiveness as the node state, where read access is less permissive than write access or read/write access). For example, if the node state is modified, any state may be maintained by a caching node. If the node state is shared, only the shared or invalid states are consistent with the node state. Since other coherent agents may have a copy of a remote cache block, if theL2 cache 36 evicts the remote cache block, the other coherent agents may be forced to evict the remote cache block as well. In some embodiments, theL2 cache 36 may force the eviction if the node state is modified, but not if the node state is shared. However, since theL2 cache 36 allows other coherent agents to cache the remote cache block in the modified state (and thus permits these coherent agents to modify the remote cache block), another coherent agent may have a copy of the remote cache block which is more up to date than the L2 cache's copy. - In such an embodiment, the
L2 cache 36 may use the WrFlush transaction to evict remote cache blocks having a modified node state. TheL2 cache 36 may initiate the WrFlush transaction on theinterconnect 22. The coherent agents may snoop the WrFlush transaction, and may invalidate any copies of the remote cache block in the coherent agents in response. Each coherent agent may signal a snoop response to theL2 cache 36 during a response portion of the WrFlush transaction. If a coherent agent has a modified copy of the cache block, that coherent agent may signal exclusive in the response portion. That coherent agent may then become responsible for transmitting the remote cache block in the data portion of the WrFlush transaction. If no coherent agent signals exclusive, theL2 cache 36 transmits the remote cache block in the data portion of the WrFlush transaction. In either case, thememory bridge 32 receives the WrFlush transaction and the remote cache block for transfer to the home node of the remote cache block. - In one embodiment, the
L2 cache 36 may use the WrFlush transaction to evict remote cache blocks to thememory bridge 32 and may use a different transaction (e.g. a write (Wr) transaction) to evict local cache blocks. Thememory controller 14 may receive the Wr transactions and may update thememory 24 with the local cache blocks received from theL2 cache 36. In some embodiments, the coherent agents do not snoop the Wr transaction. In some embodiments, thememory controller 14 may not process the WrFlush transaction. For example, thememory controller 14 may detect the WrFlush transaction and ignore the transaction, or may detect that the transaction address a remote cache block and ignore the transaction. - In some embodiments, a coherent agent may not necessarily determine if it has a modified copy of the cache block to signal exclusive in the response portion. For example, the
processors 12A-12N may maintain a set of snoop tags for the data caches, used for snooping purposes. The snoop tags may track the state of the tags in the data cache, except for the modified state. That is, the snoop tags may have invalid, shared, or exclusive states (where exclusive or modified in the data cache is represented by the exclusive state). In such embodiments, aprocessors 12A-12N may transfer the cache block for the WrFlush transaction if the processor has the cache block in either the exclusive or the modified state. - In one embodiment, the WrFlush transaction may be decoded by the snooping agents (e.g. the
processors 12A-12N) as an exclusive read transaction. Thus, the WrFlush transaction may have the same effect on the snooping agent as an exclusive read transaction would have. - It is noted that, while the WrFlush transaction is described above for transmitting a remote cache block to the
memory bridge 32, the WrFlush transaction may have other uses as well. Generally, the WrFlush transaction may be initiated by an initiating agent to transfer a cache block to a second agent, where the cache block may be transmitted by either the initiating agent or a third agent dependent upon the state of the cache block in the third agent. - FIGS. 2 and 3 are block diagrams illustrating the
L2 cache 36, theprocessor 12A, and thememory bridge 32 for example operation of the WrFlush transaction according to one embodiment of thenode 10. FIG. 2 is an example of the WrFlush transaction in which theprocessor 12A has the remote cache block in the shared state, and FIG. 3 is an example of the WrFlush transaction in which theprocessor 12A has the remote cache block in the modified state. The remote cache block is addressed by the address “A1” in the examples. - In the example of FIG. 2, the
L2 cache 36 has the remote cache block addressed by the address A1 in the modified state. In other words, the node state of the remote cache block is modified. Additionally, theprocessor 12A has a copy of the remote cache block in the shared state. TheL2 cache 36 evicts the cache block and, detecting that the cache block is a remote cache block, initiates a WrFlush transaction to transfer the remote cache block to the memory bridge 32 (arrow 40). Additionally, theprocessor 12A snoops the WrFlush transaction, and detects that the remote cache block is cached in the shared state. Theprocessor 12A invalidates the remote cache block in its cache. During a response portion of the WrFlush transaction, theprocessor 12A transmits a shared response to the L2 cache 36 (arrow 42). Since there is not an exclusive response, theL2 cache 36 transmits the remote cache block during the data portion of the transaction (arrow 44). Thememory bridge 32 receives the remote cache block, and subsequently transmits the remote cache block to the home node (not shown in FIG. 2). - In the example of FIG. 3, the
L2 cache 36 has the remote cache block addressed by the address A1 in the modified state. In other words, the node state of the remote cache block is modified. Additionally, theprocessor 12A has a copy of the remote cache block in the modified state. Theprocessor 12A's copy is more up to date than the L2 cache's copy, and thus should be transmitted to thememory bridge 32 for return to the home node. TheL2 cache 36 evicts the cache block and, detecting that the cache block is a remote cache block, initiates a WrFlush transaction to transfer the remote cache block to the memory bridge 32 (arrow 50). Additionally, theprocessor 12A snoops the WrFlush transaction, and detects that the remote cache block is cached in the modified state. Theprocessor 12A invalidates the remote cache block in its cache. During a response portion of the WrFlush transaction, theprocessor 12A transmits an exclusive response to the L2 cache 36 (arrow 52). Since there is an exclusive response, theL2 cache 36 inhibits transmission of the remote cache block during the data portion of the transaction. Instead, theprocessor 12A transmits the remote cache block during the data portion (arrow 54). Thememory bridge 32 receives the remote cache block, and subsequently transmits the remote cache block to the home node (not shown in FIG. 3). - It is noted that, in various embodiments, there may be additional coherent agents coupled to the interconnect22 (e.g.
additional processors 12A-12N, I/O bridges, I/O interfaces, etc.). If any coherent agent responds exclusive, that coherent agent may transmit the remote cache block in the data portion of the WrFlush transaction (similar to FIG. 3). If no coherent agent responds exclusive, the data portion of the WrFlush transaction may proceed as in FIG. 2. As mentioned above, in some embodiments, theprocessors 12A-12N (and/or other coherent snooping agents) may not determine whether or not a remote cache block is exclusive or modified to respond in the response phase. In such embodiments, the snooping agent may respond exclusive if the remote cache block is in the exclusive state or the modified state in the snooping agent, and may signal with the data transfer whether or not the remote cache block is modified. - It is noted that, in some embodiments, a given coherent agent may invalidate a cache block in its cache at a delayed time with regard to the snoop that causes the block to be invalidated. For example, the snoop may be queued for later invalidation, and the queue may be checked in parallel with a cache access to ensure the invalidated cache block is not used.
- Generally, a “snoop” may include receiving a transaction initiated on the
interconnect 22, and checking for the state of the cache block addressed by the transaction in response to the transaction. The result of the snoop may be signaled (e.g. during a response portion of the transaction), and a state change may be effected in the snooping agent for the cache block addressed by the transaction (if appropriate based on the transaction and the current state, according to the coherency protocol). A copy of a cache block may be modified if the copy has been changed by the caching agent from the copy that was supplied to that caching agent. - It is noted that, in some embodiments, the
L2 cache 36 may be programmable to reserve one or more locations for storing remote cache blocks. Such embodiments may inhibit selecting the reserved locations for replacement to store local cache blocks that miss in theL2 cache 36. For example, set associative embodiments may be programmable to reserve one or more ways for remote cache blocks. - It is noted that portions of the WrFlush transaction are referred to above (e.g. the response portion and the data portion). Depending on the nature of the
interconnect 22, the portions may occur in various fashions. For example, a packet-based interconnect may include a request packet to initiate the transaction (including the address A1), one or more response packets indicating the response (which may be included in the response portion) and a data packet (which may be included in the data portion). If the interconnect is a bus, the response portion and data portion may be phases on the bus (e.g. a split transaction bus may include an address phase on the address bus, a response phase on the response lines, and a data phase on the data bus). - Turning next to FIG. 4, a flowchart is shown illustrating operation of one embodiment of the
L2 cache 36 in response to determining that a cache block is to be evicted (e.g. in response to an L2 cache miss for a different cache block). The blocks in FIG. 4 are illustrated in a particular order for ease of understanding, but any order may be used. Furthermore, blocks may be performed in parallel by combinatorial logic circuitry in theL2 cache 36. Still further, blocks may be pipelined over multiple clock cycles and/or the flowchart of FIG. 4 may represent operation over a multiple clock cycles. - The
L2 cache 36 may determine if the evicted cache block is in the modified state (decision block 60). If the evicted cache block is not modified, in this embodiment, theL2 cache 36 need not perform a transaction on theinterconnect 22 irrespective of whether the evicted cache block is local or remote. Thus, if the evicted cache block is not in the modified state (decision block 60—“no” leg), theL2 cache 36 may drop the evicted cache block (block 62). - If the evicted cache block is modified (
decision block 60—“yes” leg), the L2 cache block may determine if the cache block is a remote cache block (decision block 64). In one embodiment, a remote cache block may be detectable by its address. For example, one embodiment described below with regard to FIGS. 5-9 maps addresses to nodes based on the most significant bits of address and the node number in the node. Other embodiments may detect remote cache blocks in other fashions (e.g. theL2 cache 36 may store an indication for each cache block, indicating if it is local or remote). If the cache block is local (decision block 64—“no” leg), theL2 cache 36 initiates a write transaction (Wr transaction) on theinterconnect 22 to transfer the local cache block to the memory controller 14 (block 66). This transaction may not be snooped, and thus theL2 cache 36 may transmit the cache block during the data portion of the Wr transaction. - If the cache block is remote (
decision block 64—“yes” leg), theL2 cache 36 initiates a WrFlush transaction on the interconnect 22 (block 68). TheL2 cache 36 may wait for the response portion of the WrFlush transaction and, if an exclusive response is received (decision block 70—“yes” leg), theL2 cache 36 may inhibit transmitting the cache block during the data portion. On the other hand, if an exclusive response is not received (decision block 70—“no” leg), theL2 cache 36 transmits the cache block to thememory bridge 32 during the data portion of the transaction (block 72). - Additional CC-NUMA Details, One Embodiment
- FIGS.5-9 illustrate additional details regarding one exemplary embodiment of a CC-NUMA protocol that may be employed by one embodiment of the
node 10. The embodiment of FIGS. 5-9 is merely exemplary. Numerous other implementations of CC-NUMA protocols or other distributed memory system protocols may be used in other embodiments. - Turning next to FIG. 5, a table142 is shown illustrating an exemplary set of transactions supported by one embodiment of the
interconnect 22 and a table 144 is shown illustrating an exemplary set of coherency commands supported by one embodiment of the interfaces 30. Other embodiments including subsets, supersets, or alternative sets of commands may be used. - The transactions illustrated in the table142 will next be described. An agent in the
node 10 may read a cache block (either remote or local) using the read shared (RdShd) or read exclusive (RdExc) transactions on theinterconnect 22. The RdShd transaction is used to request a shared copy of the cache block, and the RdExc transaction is used to request an exclusive copy of the cache block. If the RdShd transaction is used, and no other agent reports having a copy of the cache block during the response phase of the transaction (except for theL2 cache 36 and/or the memory controller 14), the agent may take the cache block in the exclusive state. In response to the RdExc transaction, other agents in the node invalidate their copies of the cache block (if any). Additionally, an exclusive (or modified) owner of the cache block may supply the data for the transaction in the data phase. Other embodiments may employ other mechanisms (e.g. a retry on the interconnect 22) to ensure the transfer of a modified cache block. - The write transaction (Wr) and the write invalidate transaction (WrInv) may be used by an agent to write a cache block to memory. The Wr transaction may be used by an owner having the modified state for the block, since no other copies of the block need to be invalidated. The WrInv transaction may be used by an agent that does not have exclusive ownership of the block (the agent may even have the invalid state for the block). The WrInv transaction causes other agents to invalidate any copies of the block, including modified copies. The WrInv transaction may be used by an agent that is writing the entire cache block. For example, a DMA that is writing the entire cache block with new data may use the transaction to avoid a read transaction followed by a write transaction.
- The RdKill and RdInv transactions may be used by the
memory bridge 32 in response to probes received by thenode 10 from other nodes. The RdKill and RdInv transactions cause the initiator (the memory bridge 32) to acquire exclusive access to the cache block and cause any cache agents to invalidate their copies (transferring data to the initiator similar to the RdShd and RdExc transactions). In one embodiment, the RdKill transaction also cancels a reservation established by the load-linked instruction in the MIPS instruction set, while the RdInv transaction does not. In other embodiments, a single transaction may be used for probes. In still other embodiments, there may be a probe-generated transaction that invalidates agent copies of the cache block (similar to the RdKill and RdInv transactions) and another probe-generated transaction that permits agents to retain shared copies of the cache block. - The WrFlush transaction is a write transaction which may be initiated by an agent and another agent may have an exclusive or modified copy of the block. The other agent provides the data for the WrFlush transaction, or the initiating agent provides the data if no other agent has an exclusive or modified copy of the block. The WrFlush transaction may be used, in one embodiment as described above by the
L2 cache 36. - The Nop transaction is a no-operation transaction. The Nop may be used if an agent is granted use of the interconnect22 (e.g. the address bus, in embodiments in which the
interconnect 22 is a split transaction bus) and the agent determines that it no longer has a transaction to run on theinterconnect 22. - The commands illustrated in the table144 will next be described. In the table 144, the command is shown as well as the virtual channel in which the command travels on the interfaces 30. The virtual channels may include, in the illustrated embodiment: the coherent read (CRd) virtual channel; the probe (Probe) virtual channel; the acknowledge (Ack) virtual channel; and coherent fill (CFill) virtual channel. The CRd, Probe, Ack, and CFill virtual channels are defined for the HTcc commands. There may be additional virtual channels for the standard HT commands (e.g. non-posted command (NPC) virtual channel, the posted command (PC) virtual channel, and the response (RSP) virtual channel).
- The cRdShd or cRdExc commands may be issued by the
memory bridge 32 in response to a RdShd or RdExc transactions on theinterconnect 22, respectively, to read a remote cache block not stored in the node (or, in the case of RdExc, the block may be stored in the node but in the shared state). If the cache block is stored in the node (with exclusive ownership, in the case of the RdExc transaction), the read is completed on theinterconnect 22 without any coherency command transmission by thememory bridge 32. - The Flush and Kill commands are probe commands for this embodiment. The
memory bridge 32 at the home node of a cache block may issue probe commands in response to a cRdShd or cRdExc command. Thememory bridge 32 at the home node of the cache block may also issue a probe command in response to a transaction for a local cache block, if one or more remote nodes has a copy of the cache block. The Flush command is used to request that a remote modified owner of a cache block return the cache block to the home node (and invalidate the cache block in the remote modified owner). The Kill command is used to request that a remote owner invalidate the cache block. In other embodiments, additional probe commands may be supported for other state change requests (e.g. allowing remote owners to retain a shared copy of the cache block). - The probe commands are responded to (after effecting the state changes requested by the probe commands) using either the Kill_Ack or WB commands. The Kill_Ack command is an acknowledgement that a Kill command has been processed by a receiving node. The WB command is a write back of the cache block, and is transmitted in response to the Flush command. The WB command may also be used by a node to write back a remote cache block that is being evicted from the node.
- The Fill command is the command to transfer data to a remote node that has transmitted a read command (cRdExc or cRdShd) to the home node. The Fill command is issued by the
memory bridge 32 in the home node after the probes (if any) for a cache block have completed. - Turning next to FIG. 6, a block diagram illustrating one embodiment of an address space implemented by one embodiment of the
node 10 is shown. Addresses shown in FIG. 6 are illustrated as hexadecimal digits, with an under bar (“_”) separating groups of four digits. Thus, in the embodiment illustrated in FIG. 6, 40 bits of address are supported. In other embodiments, more or fewer address bits may be supported. - In the embodiment of FIG. 6, the address space between 00—0000—0000 and 0F_FFFF_FFFF is treated as local address space. Transactions generated by agents in the local address space do not generate coherency commands to other nodes, although coherency may be enforced within the
node 10 for these addresses. That is, the local address space is not maintained coherent with other nodes. Various portions of the local address space may be memory mapped to I/O devices, HT, etc. as desired. - The address space between 40—0000—0000 and EF_FFFF_FFFF is the remote
coherent space 148. That is, the address space between 40—0000—0000 and EF_FFFF_FFFF is maintained coherent between the nodes. Each node is assigned a portion of the remote coherent space, and that node is the home node for the portion. As shown in FIG. 1, each node is programmable with a node number. The node number is equal to the most significant nibble (4 bits) of the addresses for which that node is the home node, in this embodiment. Thus, the node numbers may range from 4 to E in the embodiment shown. Other embodiments may support more or fewer node numbers, as desired. In the illustrated embodiment, each node is assigned a 64 Gigabyte (GB) portion of the memory space for which it is the home node. The size of the portion assigned to each node may be varied in other embodiments (e.g. based on the address size or other factors). - For a given coherent node, there is an aliasing between the remote coherent space for which that node is the home node and the local address space of that node. That is, corresponding addresses in the local address space and the portion of the remote coherent space for which the node is the home node access the same memory locations in the
memory 24 of the node (or are memory mapped to the same I/O devices or interfaces, etc.). For example, the node havingnode number 5 aliases theaddress space 50—0000—0000 through 5F_FFFF_FFFF to 00—0000—0000 through 0F_FFFF_FFFF respectively (arrow 146). Internode coherent accesses to thememory 24 at thenode 10 use the node-numbered address space (e.g. 50—0000—0000 to 5F_FFFF_FFFF, if the node number programmed intonode 10 is 5) to access cache blocks in thememory 24. That is agents in other nodes and agents within the node that are coherently accessing cache blocks in the memory use the remote coherent space, while access in the local address space are not maintained coherent with other nodes (even though the same cache block may be accessed). Thus the addresses are aliased, but not maintained coherent, in this embodiment. In other embodiments, the addresses in the remote coherent space and the corresponding addresses in the local address space may be maintained coherent. - A cache block is referred to as local in a node if the cache block is part of the memory assigned to the node (as mentioned above). Thus, the cache block may be local if it is accessed from the local address space or the remote coherent space, as long as the address is in the range for which the node is the home node. Similarly, a transaction on the
interconnect 22 that accesses a local cache block may be referred to as a local transaction or local access. A transaction on theinterconnect 22 that accesses a remote cache block (via the remote coherent address space outside of the portion for which the node is the home node) may be referred to as a remote transaction or a remote access. - The address space between 10—0000—0000 and 3F_FFFF_FFFF may be used for additional HT transactions (e.g. standard HT transactions) in the illustrated embodiment. Additionally, the address space between F0—0000—0000 and FF_FFFF_FFFF may be reserved in the illustrated embodiment.
- It is noted that, while the most significant nibble of the address defines which node is being accessed, other embodiments may use any other portion of the address to identify the node. Furthermore, other information in the transaction may be used to identify remote versus local transactions, in other embodiments (e.g. command type, control information transmitted in the transaction, etc.).
- Turning next to FIG. 7, a decision tree for a read transaction to a memory space address on the
interconnect 22 of anode 10 is shown for one embodiment. The decision tree may illustrate operation of thenode 10 for the read transaction for different conditions of the transaction, the state of the cache block accessed by the transaction, etc. The read transaction may, in one embodiment, include the RdShd, RdExc, RdKill, and RdInv transactions shown in the table 142 of FIG. 5. Each dot on the lines within the decision tree represents a divergence point of one or more limbs of the tree, which are labeled with the corresponding conditions. Where multiple limbs emerge from a dot, taking one limb also implies that the conditions for the other limbs are not met. In FIG. 7, the exclamation point (“!”) is used to indicate a logical NOT. Not shown in FIG. 7 is the state transition made by each coherent agent which is caching a copy of the cache block for the read transaction. If the read transaction is RdShd, the coherent agent may retain a copy of the cache block in the shared state. Otherwise, the coherent agent invalidates its copy of the cache block. - The transaction may be either local or remote, as mentioned above. For local transactions, if the transaction is uncacheable, then a read from the
memory 24 is performed (reference numeral 150). In one embodiment, the transaction may include an indication of whether or not the transaction is cacheable. If the transaction is uncacheable, it is treated as a non-coherent transaction in the present embodiment. - If the local transaction is cacheable, the operation of the
node 10 is dependent on the response provided during the response phase of the transaction. In one embodiment, each coherent agent responds with the state of the cache block in that agent. For example, each coherent agent may have an associated shared (SHD) and exclusive (EXC) signal. The agent may signal invalid state by deasserting both the SHD and EXC signals. The agent may signal shared state by asserting the SHD signal and deasserting the EXC signal. The agent may signal exclusive state (or modified state) by asserting the EXC signal and deasserting the SHD signal. The exclusive and modified states may be treated the same in the response phase in this embodiment, and the exclusive/modified owner may provide the data. The exclusive/modified owner may provide, concurrent with the data, an indication of whether the state is exclusive or modified. While each agent may have its own SHD and EXC signals in this embodiment (and the initiating agent may receive the signals from each other agent), in other embodiments a shared SHD and EXC signal may be used by all agents. - If both the SHD and EXC responses are received for the local transaction, an error has occurred (reference numeral152). The memory controller may return a fatal error indication for the read transaction, in one embodiment. If the response is exclusive (SHD deasserted, EXC asserted) the exclusive owner provides the data for the read transaction on the interconnect 22 (reference numeral 154). If the exclusive owner is the memory bridge 32 (as recorded in the remote line directory 34), then a remote node has the cache block in the modified state. The
memory bridge 32 issues a probe (Flush command) to retrieve the cache block from that remote node. Thememory bridge 32 may supply the cache block returned from the remote node as the data for the read on theinterconnect 22. - If the response is shared (SHD asserted, EXC deasserted), the local transaction is RdExc, and the
memory bridge 32 is one of the agents reporting shared, then at least one remote node may have a shared copy of the cache block. Thememory bridge 32 may initiate a probe (Kill command) to invalidate the shared copies of the cache block in the remote node(s) (reference numeral 156). In one embodiment, the data may be read from memory (or the L2 cache 36) for this case, but the transfer of the data may be delayed until the remote node(s) have acknowledged the probe. Thememory bridge 32 may signal thememory controller 14/L2 cache 36 when the acknowledgements have been received. In one embodiment, each transaction may have a transaction identifier on theinterconnect 22. Thememory bridge 32 may transmit the transaction identifier of the RdExc transaction to thememory controller 14/L2 cache 36 to indicate that the data may be transmitted. - If the response is shared, the local transaction is RdExc, and the sharing agents are local agents (i.e. the
memory bridge 32 does not report shared), then theL2 cache 36 or thememory controller 14 may supply the data, depending on whether or not there is an L2 hit for the cache block (reference numeral 158). Similarly, if the response is shared and the transaction is not RdExc, theL2 cache 36 or thememory controller 14 may supply the data dependent on whether or not there is an L2 hit for the cache block. - If the transaction is remote and uncacheable, then the
memory bridge 32 may generate a noncoherent read command on the interfaces 30 to read the data. For example, a standard HT read command may be used (reference numeral 160). If the remote transaction is cacheable and the response on theinterconnect 22 is exclusive, then the exclusive owner supplies the data for the read (reference numeral 162). If the remote transaction is cacheable, the response is not exclusive, the cache block is an L2 cache hit, and the transaction is either RdShd or the transaction is RdExc and the L2 cache has the block in the modified state, then theL2 cache 36 supplies the data for the read (reference numeral 164). Otherwise, thememory bridge 32 initiates a corresponding read command to the home node of the cache block (reference numeral 166). - Turning next to FIG. 8, a decision tree for a write transaction to a memory space address on the
interconnect 22 of anode 10 is shown for one embodiment. The decision tree may illustrate operation of the node for the write transaction for different conditions of the transaction, the state of the cache block accessed by the transaction, etc. The write transaction may, in one embodiment, include the Wr, WrInv, and WrFlush transactions shown in the table 142 of FIG. 5. Each dot on the lines within the decision tree represents a divergence point of one or more limbs of the tree, which are labeled with the corresponding conditions. Where multiple limbs emerge from a dot, taking one limb also implies that the conditions for the other limbs are not met. In FIG. 8, the exclamation point (“!”) is used to indicate a logical NOT. Not shown in FIG. 8 is the state transition made by each coherent agent which is caching a copy of the cache block for the write transaction. The coherent agent invalidates its copy of the cache block. - If the transaction is a local transaction, and the transaction is a WrInv transaction that hits in the remote line directory34 (i.e. a remote node is caching a copy of the cache block), the memory controller 14 (and the
L2 cache 36, if an L2 hit) updates with the write data (reference numeral 170). Additionally, thememory bridge 32 may generate probes to the remote nodes indicated by theremote line directory 34. The update of the memory/L2 cache may be delayed until the probes have been completed, at which time thememory bridge 32 may transmit the transaction identifier of the WrInv transaction to theL2 cache 36/memory controller 14 to permit the update. - If the local transaction is uncacheable or if the
L2 cache 36 is the master of the transaction (that is, theL2 cache 36 initiated the transaction), then thememory controller 14 updates with the data (reference numeral 172). If the local transaction is cacheable, thememory controller 14 and/or theL2 cache 36 updates with the data based on whether or not there is an L2 cache hit (and, in some embodiments, based on an L2 cache allocation indication in the transaction, which allows the source of the transaction to indicate whether or not the L2 cache allocates a cache line for an L2 cache miss) (reference numeral 174). - If the transaction is a remote transaction, the transaction is a WrFlush transaction, and the response to the transaction is exclusive, the exclusive owner supplies the data (reference numeral176). If the remote WrFlush transaction results in a non-exclusive response (shared or invalid), the
L2 cache 36 supplies the data of the WrFlush transaction. In one embodiment, theL2 cache 36 retains the state of the node as recorded in the home node, and theL2 cache 36 uses the WrFlush transaction to evict a remote cache block which is in the modified state in the node. Thus, if another agent has the cache block in the exclusive state, that agent may have a more recent copy of the cache block that should be returned to the home node. Otherwise, theL2 cache 36 supplies the block to be returned to the home node (reference numeral 182). In either case, thememory bridge 32 may capture the WrFlush transaction and data, and may perform a WB command to return the cache block to the home node. - If the remote transaction is not a WrFlush transaction, and is not cache coherent, the
memory bridge 32 receives the write transaction and performs a noncoherent Wr command (e.g. a standard HT write) to transmit the cache block to the home node (reference numeral 180). If the remote transaction is not a WrFlush transaction, is cache coherent, and is an L2 hit, theL2 cache 36 may update with the data (reference numeral 182). - Turning next to FIG. 9, a block diagram illustrating operation of one embodiment of the
memory bridge 32 in response to various coherency commands received from theinterface circuits 20A-20C is shown. The received command is shown in an oval. Commands initiated by thememory bridge 32 in response to the received command (and the state of the affected cache block as indicated in the remote line directory 34) are shown in solid boxes. Dotted boxes are commands received by thememory bridge 32 in response to the commands transmitted in the preceding solid boxes. The cache block affected by a command is shown in parentheses after the command. - In one embodiment, the
remote line directory 34 may be accessed in response to a transaction on theinterconnect 22. In such an embodiment, thememory bridge 32 may initiate a transaction on theinterconnect 22 in response to certain coherent commands in order to retrieve the remote line directory 34 (as well as to affect any state changes in the coherent agents coupled to theinterconnect 22, if applicable). In other embodiments, thememory bridge 32 may be configured to read theremote line directory 34 prior to generating a transaction on theinterconnect 22, and may conditionally generate a transaction if needed based on the state of theremote line directory 34 for the requested cache block. Additionally, in one embodiment, theremote line directory 34 may maintain the remote state for a subset of the local cache blocks that are shareable remotely (e.g. a subset of the portion of the remotecoherent space 148 that is assigned to the local node). If a cache block is requested by a remote node using a coherency command and there is no entry in theremote line directory 34 for the cache block, then a victim cache block may be replaced in the remote line directory 34 (and probes may be generated to invalidate the victim cache block in remote nodes). In other embodiments, theremote line directory 34 may be configured to track the state of each cache block in the portion of the remotecoherent space 148 that is assigned to the local node. In such embodiments, operations related to the victim cache blocks may be omitted from FIG. 9. - For a cRdShd command for cache block “A” received by the memory bridge32 (reference numeral 190), the
memory bridge 32 may generate a RdShd transaction on theinterconnect 22. Based on the remote line directory (RLD) state for the cache block A, a number of operations may occur. If the RLD state is shared, or invalid and there is an entry available for allocation without requiring a victim cache block to be evicted (“RLD empty” in FIG. 9), then thememory bridge 32 may transmit a fill command to the remote node with the data supplied to thememory bridge 32 in response to the RdShd transaction on the interconnect 22 (reference numeral 192). On the other hand, if the RLD state is invalid and an eviction of a victim block is used to free an RLD entry for cache block A, then thememory bridge 32 may transmit probes to the remote nodes having copies of the victim cache block. If the victim cache block is shared, thememory bridge 32 may transmit a Kill command (or commands, if multiple nodes are sharing the victim cache block) for the victim block (reference numeral 194). The remote nodes respond with Kill_Ack commands for the victim block (reference numeral 196). If the victim block is modified, thememory bridge 32 may transmit a Flush command to the remote node having the modified state (reference numeral 198). The remote node may return the modified block with a WB command (reference numeral 200). In either case of evicting a victim block, thememory bridge 32 may, in parallel, generate a Fill command for the cache block A (reference numeral 92, via arrow 202). Finally, if the RLD state is modified for the cache block A, thememory bridge 32 may generate a Flush command for the cache block A to the remote node (reference numeral 204), which responds with a WB command and the cache block A (reference numeral 206). Thememory bridge 32 may then transmit the Fill command with the cache block A provided via the write back command (reference numeral 192). - In response to a cRdExc command for a cache block A (reference numeral210), operation may be similar to the cRdShd case for some RLD states. Similar to the cRdShd case, the
memory bridge 32 may initiate a RdExc transaction on theinterconnect 22 in response to the cRdExc command. Similar to the cRdShd case, if the RLD is invalid and no eviction of a victim cache block is needed in the RLD to allocate an entry for the cache block A, then thememory bridge 32 may supply the cache block supplied on theinterconnect 22 for the RdExc transaction in a fill command to the remote node (reference numeral 212). Additionally, if the RLD state is invalid for the cache block A and a victim cache block is evicted from theRLD 34, thememory bridge 32 may operate in a similar fashion to the cRdShd case (reference numerals arrow 222 for the shared case of the victim block andreference numerals arrow 222 for the modified case of the victim block). If the RLD state is modified for the cache block A, thememory bridge 32 may operate in a similar fashion to the cRdShd case (reference numerals 224 and 226). If the RLD state is shared for the cache block A, thememory bridge 32 may generate Kill commands for each remote sharing node (reference numeral 228). Thememory bridge 32 may wait for the Kill_Ack commands from the remote sharing nodes (reference numeral 230), and then transmit the Fill command with the cache block A provided on theinterconnect 22 in response to the RdExc transaction (reference numeral 212). - In response to a Wr command to the cache block A, the
memory bridge 32 may generate a Wr transaction on the interconnect 22 (reference numeral 240). If the RLD state is invalid for the cache block A, thememory bridge 32 may transmit the write data on theinterconnect 22 and the Wr command is complete (reference numeral 242). If the RLD state is shared for the cache block A, thememory bridge 32 may generate Kill commands to each remote sharing node (reference numeral 244) and collect the Kill_Ack commands from those remote nodes (reference numeral 246) in addition to transmitting the data on theinterconnect 22. If the RLD state is modified for a remote node, thememory bridge 32 may generate a Flush command to the remote node (reference numeral 248) and receive the WB command from the remote node (reference numeral 250). In one embodiment, thememory bridge 32 may delay transmitting the write data on theinterconnect 22 until the WB command or Kill_Ack commands are received (although the data returned with the WB command may be dropped by the memory bridge 32). - The above commands are received by the
memory bridge 32 for cache blocks for which thenode 10 including thememory bridge 32 is the home node. Thememory bridge 32 may also receive Flush commands or Kill commands for cache blocks for which thenode 10 is a remote node. In response to a Flush command to the cache block A (reference numeral 260), thememory bridge 32 may initiate a RdInv transaction on theinterconnect 22. If the local state of the cache block is modified, thememory bridge 32 may transmit a WB command to the home node, with the cache block supplied on theinterconnect 22 in response to the Rdlnv transaction (reference numeral 262). If the local state of the cache block is not modified, thememory bridge 32 may not respond to the Flush command (reference numeral 264). In this case, the node may already have transmitted a WB command to the home node (e.g. in response to evicting the cache block locally). In response to a Kill command to the cache block A (reference numeral 270), thememory bridge 32 may initiate a RdKill transaction on theinterconnect 22. Thememory bridge 32 may respond to the Kill command with a Kill_Ack command (reference numeral 272). - In one embodiment, the
memory bridge 32 may also be configured to receive a non-cacheable read (RdNC) command (e.g. corresponding to a standard HT read) (reference numeral 280). In response, thememory bridge 32 may initiate a RdShd transaction on theinterconnect 22. If the RLD state is modified for the cache block including the data to be read, thememory bridge 32 may transmit a Flush command to the remote node having the modified cache block (reference numeral 282), and may receive the WB command from the remote node (reference numeral 284). Additionally, thememory bridge 32 may supply data received on theinterconnect 22 in response to the RdShd transaction as a read response (RSP) to the requesting node (reference numeral 286). - Computer Accessible Medium
- Turning next to FIG. 10, a block diagram of a computer
accessible medium 300 including one or more data structures representative of the circuitry included in thenode 10 is shown. Generally speaking, a computer accessible medium may include storage media such as magnetic or optical media, e.g., disk, CD-ROM, or DVD-ROM, volatile or non-volatile memory media such as RAM (e.g. SDRAM, RDRAM, SRAM, etc.), ROM, etc., as well as media accessible via transmission media or signals such as electrical, electromagnetic, or digital signals, conveyed via a communication medium such as a network and/or a wireless link. - Generally, the data structure(s) of the circuitry on the computer
accessible medium 300 may be read by a program and used, directly or indirectly, to fabricate the hardware comprising the circuitry. For example, the data structure(s) may include one or more behavioral-level descriptions or register-transfer level (RTL) descriptions of the hardware functionality in a high level design language (HDL) such as Verilog or VHDL. The description(s) may be read by a synthesis tool which may synthesize the description to produce one or more netlist(s) comprising lists of gates from a synthesis library. The netlist(s) comprise a set of gates which also represent the functionality of the hardware comprising the circuitry. The netlist(s) may then be placed and routed to produce one or more data set(s) describing geometric shapes to be applied to masks. The masks may then be used in various semiconductor fabrication steps to produce a semiconductor circuit or circuits corresponding to the circuitry. Alternatively, the data structure(s) on computeraccessible medium 300 may be the netlist(s) (with or without the synthesis library) or the data set(s), as desired. In yet another alternative, the data structures may comprise the output of a schematic program, or netlist(s) or data set(s) derived therefrom. - While computer
accessible medium 300 includes a representation of thenode 10, other embodiments may include a representation of any portion of the node 10 (e.g. processors 12A-12N,memory controller 14,L2 cache 36,interconnect 22,memory bridge 32,remote line directory 34,switch 18, interface circuits 22A-22C, etc.). - Numerous variations and modifications will become apparent to those skilled in the art once the above disclosure is fully appreciated. It is intended that the following claims be interpreted to embrace all such variations and modifications.
Claims (22)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/413,916 US20030229676A1 (en) | 2002-05-15 | 2003-04-15 | Command to transfer data from node state agent to memory bridge |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US38074002P | 2002-05-15 | 2002-05-15 | |
US10/270,028 US7266587B2 (en) | 2002-05-15 | 2002-10-11 | System having interfaces, switch, and memory bridge for CC-NUMA operation |
US10/413,916 US20030229676A1 (en) | 2002-05-15 | 2003-04-15 | Command to transfer data from node state agent to memory bridge |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/270,028 Continuation-In-Part US7266587B2 (en) | 2002-05-15 | 2002-10-11 | System having interfaces, switch, and memory bridge for CC-NUMA operation |
Publications (1)
Publication Number | Publication Date |
---|---|
US20030229676A1 true US20030229676A1 (en) | 2003-12-11 |
Family
ID=29272888
Family Applications (7)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/270,028 Expired - Fee Related US7266587B2 (en) | 2002-05-15 | 2002-10-11 | System having interfaces, switch, and memory bridge for CC-NUMA operation |
US10/413,916 Abandoned US20030229676A1 (en) | 2002-05-15 | 2003-04-15 | Command to transfer data from node state agent to memory bridge |
US10/414,363 Expired - Lifetime US6948035B2 (en) | 2002-05-15 | 2003-04-15 | Data pend mechanism |
US10/413,917 Expired - Fee Related US6988168B2 (en) | 2002-05-15 | 2003-04-15 | Cache programmable to partition ways to agents and/or local/remote blocks |
US10/435,189 Expired - Fee Related US7343456B2 (en) | 2002-05-15 | 2003-05-09 | Load-linked/store conditional mechanism in a CC-NUMA system |
US10/439,343 Expired - Fee Related US7340546B2 (en) | 2002-05-15 | 2003-05-15 | Addressing scheme supporting fixed local addressing and variable global addressing |
US11/891,954 Expired - Fee Related US7469275B2 (en) | 2002-05-15 | 2007-08-14 | System having interfaces, switch, and memory bridge for CC-NUMA operation |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/270,028 Expired - Fee Related US7266587B2 (en) | 2002-05-15 | 2002-10-11 | System having interfaces, switch, and memory bridge for CC-NUMA operation |
Family Applications After (5)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/414,363 Expired - Lifetime US6948035B2 (en) | 2002-05-15 | 2003-04-15 | Data pend mechanism |
US10/413,917 Expired - Fee Related US6988168B2 (en) | 2002-05-15 | 2003-04-15 | Cache programmable to partition ways to agents and/or local/remote blocks |
US10/435,189 Expired - Fee Related US7343456B2 (en) | 2002-05-15 | 2003-05-09 | Load-linked/store conditional mechanism in a CC-NUMA system |
US10/439,343 Expired - Fee Related US7340546B2 (en) | 2002-05-15 | 2003-05-15 | Addressing scheme supporting fixed local addressing and variable global addressing |
US11/891,954 Expired - Fee Related US7469275B2 (en) | 2002-05-15 | 2007-08-14 | System having interfaces, switch, and memory bridge for CC-NUMA operation |
Country Status (4)
Country | Link |
---|---|
US (7) | US7266587B2 (en) |
EP (1) | EP1363196B1 (en) |
AT (1) | ATE309574T1 (en) |
DE (1) | DE60207210T2 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040215473A1 (en) * | 2003-04-24 | 2004-10-28 | Sun Microsystems, Inc. | Simultaneous global transaction and local transaction management in an application server |
US20050015353A1 (en) * | 2003-07-14 | 2005-01-20 | Sun Microsystems, Inc. | Read/write lock transaction manager freezing |
US20060003439A1 (en) * | 2004-07-02 | 2006-01-05 | Ismagilov Rustem F | Microfluidic system |
US20060143406A1 (en) * | 2004-12-27 | 2006-06-29 | Chrysos George Z | Predictive early write-back of owned cache blocks in a shared memory computer system |
Families Citing this family (83)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7752281B2 (en) * | 2001-11-20 | 2010-07-06 | Broadcom Corporation | Bridges performing remote reads and writes as uncacheable coherent operations |
US7266587B2 (en) * | 2002-05-15 | 2007-09-04 | Broadcom Corporation | System having interfaces, switch, and memory bridge for CC-NUMA operation |
US8244990B2 (en) * | 2002-07-16 | 2012-08-14 | Oracle America, Inc. | Obstruction-free synchronization for shared data structures |
US7293143B1 (en) | 2002-09-24 | 2007-11-06 | Sun Microsystems, Inc. | Efficient non-blocking k-compare-single-swap operation |
US7290093B2 (en) * | 2003-01-07 | 2007-10-30 | Intel Corporation | Cache memory to support a processor's power mode of operation |
US7606978B2 (en) * | 2003-04-11 | 2009-10-20 | Sun Microsystems, Inc. | Multi-node computer system implementing global access state dependent transactions |
US20050010615A1 (en) * | 2003-04-11 | 2005-01-13 | Sun Microsystems, Inc. | Multi-node computer system implementing memory-correctable speculative proxy transactions |
WO2004092968A2 (en) * | 2003-04-11 | 2004-10-28 | Sun Microsystems, Inc. | Multi-node system with global access states |
US7398359B1 (en) * | 2003-04-30 | 2008-07-08 | Silicon Graphics, Inc. | System and method for performing memory operations in a computing system |
US7114042B2 (en) * | 2003-05-22 | 2006-09-26 | International Business Machines Corporation | Method to provide atomic update primitives in an asymmetric heterogeneous multiprocessor environment |
US20040267919A1 (en) * | 2003-06-30 | 2004-12-30 | International Business Machines Corporation | Method and system for providing server management peripheral caching using a shared bus |
US7107367B1 (en) * | 2003-08-12 | 2006-09-12 | Advanced Micro Devices, Inc. | Method for efficient buffer tag allocation |
US7539190B2 (en) * | 2004-01-05 | 2009-05-26 | Topside Research, Llc | Multicasting in a shared address space |
US7478211B2 (en) * | 2004-01-09 | 2009-01-13 | International Business Machines Corporation | Maintaining consistency for remote copy using virtualization |
US20050154786A1 (en) * | 2004-01-09 | 2005-07-14 | International Business Machines Corporation | Ordering updates in remote copying of data |
US7257679B2 (en) * | 2004-10-01 | 2007-08-14 | Advanced Micro Devices, Inc. | Sharing monitored cache lines across multiple cores |
US7970980B2 (en) * | 2004-12-15 | 2011-06-28 | International Business Machines Corporation | Method and apparatus for accessing memory in a computer system architecture supporting heterogeneous configurations of memory structures |
US20060161919A1 (en) * | 2004-12-23 | 2006-07-20 | Onufryk Peter Z | Implementation of load linked and store conditional operations |
US7258628B2 (en) * | 2005-01-10 | 2007-08-21 | Nelson Precision Casting Co., Ltd. | Intensified structure for connecting a golf club head body with a striking plate |
US20060179253A1 (en) * | 2005-02-10 | 2006-08-10 | International Business Machines Corporation | Data processing system, method and interconnect fabric that protect ownership transfer with a protection window extension |
US7483428B2 (en) * | 2005-02-10 | 2009-01-27 | International Business Machines Corporation | Data processing system, method and interconnect fabric supporting a node-only broadcast |
US7474658B2 (en) * | 2005-02-10 | 2009-01-06 | International Business Machines Corporation | Data processing system, method and interconnect fabric supporting concurrent operations of varying broadcast scope |
US7496710B1 (en) | 2005-04-01 | 2009-02-24 | Sun Microsystems, Inc. | Reducing resource consumption by ineffective write operations |
JP4362454B2 (en) * | 2005-04-07 | 2009-11-11 | 富士通株式会社 | Cache coherence management device and cache coherence management method |
US7366848B1 (en) * | 2005-06-02 | 2008-04-29 | Sun Microsystems, Inc. | Reducing resource consumption by ineffective write operations in a shared memory system |
US7480771B2 (en) * | 2005-08-17 | 2009-01-20 | Sun Microsystems, Inc. | Conditional synchronization mechanisms allowing multiple store operations to become visible while a flagged memory location is owned and remains unchanged |
US7680989B2 (en) | 2005-08-17 | 2010-03-16 | Sun Microsystems, Inc. | Instruction set architecture employing conditional multistore synchronization |
US7805560B2 (en) * | 2005-08-31 | 2010-09-28 | Ati Technologies Inc. | Methods and apparatus for translating messages in a computing system |
US7707361B2 (en) * | 2005-11-17 | 2010-04-27 | Apple Inc. | Data cache block zero implementation |
US7756943B1 (en) | 2006-01-26 | 2010-07-13 | Symantec Operating Corporation | Efficient data transfer between computers in a virtual NUMA system using RDMA |
US7702743B1 (en) * | 2006-01-26 | 2010-04-20 | Symantec Operating Corporation | Supporting a weak ordering memory model for a virtual physical address space that spans multiple nodes |
US7434124B2 (en) * | 2006-03-28 | 2008-10-07 | National Instruments Corporation | Reduced pattern memory in digital test equipment |
JP5137171B2 (en) * | 2006-07-24 | 2013-02-06 | ルネサスエレクトロニクス株式会社 | Data processing device |
US8205024B2 (en) * | 2006-11-16 | 2012-06-19 | International Business Machines Corporation | Protecting ownership transfer with non-uniform protection windows |
US8255577B2 (en) * | 2007-04-26 | 2012-08-28 | Hewlett-Packard Development Company, L.P. | I/O forwarding technique for multi-interrupt capable devices |
US7882327B2 (en) * | 2007-07-31 | 2011-02-01 | Advanced Micro Devices, Inc. | Communicating between partitions in a statically partitioned multiprocessing system |
US20090320036A1 (en) * | 2008-06-19 | 2009-12-24 | Joan Marie Ries | File System Object Node Management |
US8019920B2 (en) * | 2008-10-01 | 2011-09-13 | Hewlett-Packard Development Company, L.P. | Method to improve operating performance of a computing device |
US20100106874A1 (en) * | 2008-10-28 | 2010-04-29 | Charles Dominguez | Packet Filter Optimization For Network Interfaces |
US7970976B2 (en) * | 2009-03-01 | 2011-06-28 | Qualcomm Incorporated | Remote memory access using reversible host/client interface |
US8108650B2 (en) * | 2009-05-29 | 2012-01-31 | Apple Inc. | Translation lookaside buffer (TLB) with reserved areas for specific sources |
US20100332763A1 (en) * | 2009-06-30 | 2010-12-30 | International Business Machines Corporation | Apparatus, system, and method for cache coherency elimination |
US8510512B2 (en) * | 2009-08-21 | 2013-08-13 | International Business Machines Corporation | Memory coherence directory supporting remotely sourced requests of nodal scope |
US8601242B2 (en) * | 2009-12-18 | 2013-12-03 | Intel Corporation | Adaptive optimized compare-exchange operation |
JP5481669B2 (en) * | 2010-08-02 | 2014-04-23 | 株式会社日立製作所 | Cache control method, node device, manager device, and computer system |
US20120166739A1 (en) * | 2010-12-22 | 2012-06-28 | Andes Technology Corporation | Memory module and method for atomic operations in a multi-level memory structure |
US10360150B2 (en) | 2011-02-14 | 2019-07-23 | Suse Llc | Techniques for managing memory in a multiprocessor architecture |
JP2012252558A (en) * | 2011-06-03 | 2012-12-20 | Sony Corp | Nonvolatile memory, memory controller, nonvolatile memory accessing method, and program |
US8868843B2 (en) | 2011-11-30 | 2014-10-21 | Advanced Micro Devices, Inc. | Hardware filter for tracking block presence in large caches |
US10229221B1 (en) * | 2012-03-01 | 2019-03-12 | EMC IP Holding Company LLC | Techniques for cache updates based on quality of service |
US9238770B2 (en) | 2012-03-29 | 2016-01-19 | Kraton Polymers U.S. Llc | Low viscosity synthetic cement |
US9135174B2 (en) * | 2012-11-27 | 2015-09-15 | International Business Machines Corporation | Coherent attached processor proxy supporting master parking |
US9442852B2 (en) | 2012-11-27 | 2016-09-13 | International Business Machines Corporation | Programmable coherent proxy for attached processor |
US9069674B2 (en) | 2012-11-27 | 2015-06-30 | International Business Machines Corporation | Coherent proxy for attached processor |
US9021211B2 (en) | 2013-01-11 | 2015-04-28 | International Business Machines Corporation | Epoch-based recovery for coherent attached processor proxy |
US8990513B2 (en) | 2013-01-11 | 2015-03-24 | International Business Machines Corporation | Accelerated recovery for snooped addresses in a coherent attached processor proxy |
US8938587B2 (en) | 2013-01-11 | 2015-01-20 | International Business Machines Corporation | Data recovery for coherent attached processor proxy |
US9606922B2 (en) | 2013-03-01 | 2017-03-28 | International Business Machines Corporation | Selection of post-request action based on combined response and input from the request source |
US9323676B2 (en) * | 2013-03-05 | 2016-04-26 | International Business Machines Corporation | Non-data inclusive coherent (NIC) directory for cache |
US9053035B1 (en) * | 2013-11-25 | 2015-06-09 | Freescale Semiconductor, Inc. | Multi-threaded system for performing atomic binary translations |
US20160055100A1 (en) * | 2014-08-19 | 2016-02-25 | Advanced Micro Devices, Inc. | System and method for reverse inclusion in multilevel cache hierarchy |
EP3066572B1 (en) * | 2014-12-14 | 2020-02-19 | VIA Alliance Semiconductor Co., Ltd. | Cache memory budgeted by chunks based on memory access type |
US9652398B2 (en) * | 2014-12-14 | 2017-05-16 | Via Alliance Semiconductor Co., Ltd. | Cache replacement policy that considers memory access type |
WO2016097806A1 (en) * | 2014-12-14 | 2016-06-23 | Via Alliance Semiconductor Co., Ltd. | Fully associative cache memory budgeted by memory access type |
US9811468B2 (en) * | 2014-12-14 | 2017-11-07 | Via Alliance Semiconductor Co., Ltd. | Set associative cache memory with heterogeneous replacement policy |
EP3066571B1 (en) | 2014-12-14 | 2018-06-13 | VIA Alliance Semiconductor Co., Ltd. | Cache memory budgeted by ways on memory access type |
US9639276B2 (en) * | 2015-03-27 | 2017-05-02 | Intel Corporation | Implied directory state updates |
US9626310B2 (en) * | 2015-08-25 | 2017-04-18 | Atmel Corporation | Microcontroller architecture with access stealing |
US20170083331A1 (en) * | 2015-09-19 | 2017-03-23 | Microsoft Technology Licensing, Llc | Memory synchronization in block-based processors |
US9852084B1 (en) | 2016-02-05 | 2017-12-26 | Apple Inc. | Access permissions modification |
US10963409B2 (en) * | 2016-08-19 | 2021-03-30 | Arm Limited | Interconnect circuitry and a method of operating such interconnect circuitry |
US10613979B2 (en) * | 2017-11-30 | 2020-04-07 | International Business Machines Corporation | Accelerator memory coherency with single state machine |
EP3531293A1 (en) * | 2018-02-27 | 2019-08-28 | BAE SYSTEMS plc | Computing system operating a reflective memory network |
EP3759607A1 (en) * | 2018-02-27 | 2021-01-06 | BAE SYSTEMS plc | Computing system operating a reflective memory network |
US10503643B1 (en) * | 2018-07-11 | 2019-12-10 | Qualcomm Incorporated | Cache coherence with functional address apertures |
US11068407B2 (en) | 2018-10-26 | 2021-07-20 | International Business Machines Corporation | Synchronized access to data in shared memory by protecting the load target address of a load-reserve instruction |
US10884740B2 (en) | 2018-11-08 | 2021-01-05 | International Business Machines Corporation | Synchronized access to data in shared memory by resolving conflicting accesses by co-located hardware threads |
US11200168B2 (en) | 2018-12-10 | 2021-12-14 | International Business Machines Corporation | Caching data from remote memories |
US11119781B2 (en) | 2018-12-11 | 2021-09-14 | International Business Machines Corporation | Synchronized access to data in shared memory by protecting the load target address of a fronting load |
US11016913B1 (en) * | 2020-03-30 | 2021-05-25 | Apple Inc. | Inter cluster snoop latency reduction |
US11106608B1 (en) | 2020-06-22 | 2021-08-31 | International Business Machines Corporation | Synchronizing access to shared memory by extending protection for a target address of a store-conditional request |
US11693776B2 (en) | 2021-06-18 | 2023-07-04 | International Business Machines Corporation | Variable protection window extension for a target address of a store-conditional request |
US20230333856A1 (en) * | 2022-04-18 | 2023-10-19 | Cadence Design Systems, Inc | Load-Store Unit Dual Tags and Replays |
Citations (41)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4433378A (en) * | 1981-09-28 | 1984-02-21 | Western Digital | Chip topography for MOS packet network interface circuit |
US4463424A (en) * | 1981-02-19 | 1984-07-31 | International Business Machines Corporation | Method for dynamically allocating LRU/MRU managed memory among concurrent sequential processes |
US4760571A (en) * | 1984-07-25 | 1988-07-26 | Siegfried Schwarz | Ring network for communication between one chip processors |
US5640399A (en) * | 1993-10-20 | 1997-06-17 | Lsi Logic Corporation | Single chip network router |
US5668809A (en) * | 1993-10-20 | 1997-09-16 | Lsi Logic Corporation | Single chip network hub with dynamic window filter |
US5778414A (en) * | 1996-06-13 | 1998-07-07 | Racal-Datacom, Inc. | Performance enhancing memory interleaver for data frame processing |
US5778438A (en) * | 1995-12-06 | 1998-07-07 | Intel Corporation | Method and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests |
US5802287A (en) * | 1993-10-20 | 1998-09-01 | Lsi Logic Corporation | Single chip universal protocol multi-function ATM network interface |
US5829025A (en) * | 1996-12-17 | 1998-10-27 | Intel Corporation | Computer system and method of allocating cache memories in a multilevel cache hierarchy utilizing a locality hint within an instruction |
US5887187A (en) * | 1993-10-20 | 1999-03-23 | Lsi Logic Corporation | Single chip network adapter apparatus |
US5893150A (en) * | 1996-07-01 | 1999-04-06 | Sun Microsystems, Inc. | Efficient allocation of cache memory space in a computer system |
US5908468A (en) * | 1997-10-24 | 1999-06-01 | Advanced Micro Devices, Inc. | Data transfer network on a chip utilizing a multiple traffic circle topology |
US5914955A (en) * | 1993-10-20 | 1999-06-22 | Lsi Logic Corporation | Switched network hub on a chip |
US5974508A (en) * | 1992-07-31 | 1999-10-26 | Fujitsu Limited | Cache memory system and method for automatically locking cache entries to prevent selected memory items from being replaced |
US6018763A (en) * | 1997-05-28 | 2000-01-25 | 3Com Corporation | High performance shared memory for a bridge router supporting cache coherency |
US6021451A (en) * | 1994-05-20 | 2000-02-01 | Intel Corporation | Method and apparatus for maintaining transaction ordering and arbitrating in a bus bridge |
US6092137A (en) * | 1997-11-26 | 2000-07-18 | Industrial Technology Research Institute | Fair data bus arbitration system which assigns adjustable priority values to competing sources |
US6098064A (en) * | 1998-05-22 | 2000-08-01 | Xerox Corporation | Prefetching and caching documents according to probability ranked need S list |
US6111859A (en) * | 1997-01-16 | 2000-08-29 | Advanced Micro Devices, Inc. | Data transfer network on a computer chip utilizing combined bus and ring topologies |
US6128677A (en) * | 1997-10-15 | 2000-10-03 | Intel Corporation | System and method for improved transfer of data between multiple processors and I/O bridges |
US6151662A (en) * | 1997-12-02 | 2000-11-21 | Advanced Micro Devices, Inc. | Data transaction typing for improved caching and prefetching characteristics |
US6157623A (en) * | 1997-02-14 | 2000-12-05 | Advanced Micro Devices, Inc. | Apparatus and method for selectively outputting data using a MAC layer interface or a PCI bus interface |
US6202125B1 (en) * | 1996-11-25 | 2001-03-13 | Intel Corporation | Processor-cache protocol using simple commands to implement a range of cache configurations |
US6202129B1 (en) * | 1998-03-31 | 2001-03-13 | Intel Corporation | Shared cache structure for temporal and non-temporal information using indicative bits |
US6209020B1 (en) * | 1996-09-20 | 2001-03-27 | Nortel Networks Limited | Distributed pipeline memory architecture for a computer system with even and odd pids |
US6215497B1 (en) * | 1998-08-12 | 2001-04-10 | Monolithic System Technology, Inc. | Method and apparatus for maximizing the random access bandwidth of a multi-bank DRAM in a computer graphics system |
US6262594B1 (en) * | 1999-11-05 | 2001-07-17 | Ati International, Srl | Apparatus and method for configurable use of groups of pads of a system on chip |
US6266797B1 (en) * | 1997-01-16 | 2001-07-24 | Advanced Micro Devices, Inc. | Data transfer network on a computer chip using a re-configurable path multiple ring topology |
US6266743B1 (en) * | 1999-02-26 | 2001-07-24 | International Business Machines Corporation | Method and system for providing an eviction protocol within a non-uniform memory access system |
US6269427B1 (en) * | 1999-03-18 | 2001-07-31 | International Business Machines Corporation | Multiple load miss handling in a cache memory system |
US6279087B1 (en) * | 1997-12-22 | 2001-08-21 | Compaq Computer Corporation | System and method for maintaining coherency and improving performance in a bus bridge supporting write posting operations |
US6321309B1 (en) * | 1998-05-19 | 2001-11-20 | Lsi Logic Corporation | Memory arbitration scheme with circular sequence register |
US6332179B1 (en) * | 1999-08-19 | 2001-12-18 | International Business Machines Corporation | Allocation for back-to-back misses in a directory based cache |
US6349365B1 (en) * | 1999-10-08 | 2002-02-19 | Advanced Micro Devices, Inc. | User-prioritized cache replacement |
US6366583B2 (en) * | 1996-08-07 | 2002-04-02 | Cisco Technology, Inc. | Network router integrated onto a silicon chip |
US6373846B1 (en) * | 1996-03-07 | 2002-04-16 | Lsi Logic Corporation | Single chip networking device with enhanced memory access co-processor |
US6378048B1 (en) * | 1998-11-12 | 2002-04-23 | Intel Corporation | “SLIME” cache coherency system for agents with multi-layer caches |
US6438651B1 (en) * | 1999-11-01 | 2002-08-20 | International Business Machines Corporation | Method, system, and program for managing requests to a cache using flags to queue and dequeue data in a buffer |
US6519685B1 (en) * | 1999-12-22 | 2003-02-11 | Intel Corporation | Cache states for multiprocessor cache coherency protocols |
US6574708B2 (en) * | 2001-05-18 | 2003-06-03 | Broadcom Corporation | Source controlled cache allocation |
US6681293B1 (en) * | 2000-08-25 | 2004-01-20 | Silicon Graphics, Inc. | Method and cache-coherence system allowing purging of mid-level cache entries without purging lower-level cache entries |
Family Cites Families (150)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3693165A (en) * | 1971-06-29 | 1972-09-19 | Ibm | Parallel addressing of a storage hierarchy in a data processing system using virtual addressing |
AT354159B (en) * | 1975-02-10 | 1979-12-27 | Siemens Ag | ASSOCIATIVE STORAGE WITH SEPARATELY ASSOCIATED AREAS |
US4371929A (en) * | 1980-05-05 | 1983-02-01 | Ibm Corporation | Multiprocessor system with high density memory set architecture including partitionable cache store interface to shared disk drive memory |
US4513367A (en) | 1981-03-23 | 1985-04-23 | International Business Machines Corporation | Cache locking controls in a multiprocessor |
US4453212A (en) * | 1981-07-13 | 1984-06-05 | Burroughs Corporation | Extended address generating apparatus and method |
US4575792A (en) * | 1982-03-31 | 1986-03-11 | Honeywell Information Systems Inc. | Shared interface apparatus for testing the memory sections of a cache unit |
US4511994A (en) * | 1982-09-27 | 1985-04-16 | Control Data Corporation | Multi-group LRU resolver |
US4807115A (en) * | 1983-10-07 | 1989-02-21 | Cornell Research Foundation, Inc. | Instruction issuing mechanism for processors with multiple functional units |
US4654778A (en) * | 1984-06-27 | 1987-03-31 | International Business Machines Corporation | Direct parallel path for storage accesses unloading common system path |
US4633440A (en) * | 1984-12-31 | 1986-12-30 | International Business Machines | Multi-port memory chip in a hierarchical memory |
EP0239081B1 (en) * | 1986-03-26 | 1995-09-06 | Hitachi, Ltd. | Pipelined data processor capable of decoding and executing plural instructions in parallel |
CA1286784C (en) | 1986-08-27 | 1991-07-23 | Joseph A. Petolino, Jr. | Cache storage queue |
CH670715A5 (en) | 1986-10-03 | 1989-06-30 | Bbc Brown Boveri & Cie | |
JPH0673114B2 (en) * | 1987-03-31 | 1994-09-14 | 日本電気株式会社 | Cash controller |
JPH01503786A (en) * | 1987-06-25 | 1989-12-21 | イビイ イスティテュト ビオキミコ イタリアノ ジョバンニ ロレンツィニ ソチエタ ペル アツィオニ | Prostaglandin derivatives, their production methods and pharmaceutical compositions containing them |
US5025366A (en) | 1988-01-20 | 1991-06-18 | Advanced Micro Devices, Inc. | Organization of an integrated cache unit for flexible usage in cache system design |
JPH0727492B2 (en) | 1988-01-21 | 1995-03-29 | 三菱電機株式会社 | Buffer storage |
US4996641A (en) * | 1988-04-15 | 1991-02-26 | Motorola, Inc. | Diagnostic mode for a cache |
US5317716A (en) * | 1988-08-16 | 1994-05-31 | International Business Machines Corporation | Multiple caches using state information indicating if cache line was previously modified and type of access rights granted to assign access rights to cache line |
US5163142A (en) * | 1988-10-28 | 1992-11-10 | Hewlett-Packard Company | Efficient cache write technique through deferred tag modification |
US5067069A (en) | 1989-02-03 | 1991-11-19 | Digital Equipment Corporation | Control of multiple functional units with parallel operation in a microcoded execution unit |
US5125083A (en) * | 1989-02-03 | 1992-06-23 | Digital Equipment Corporation | Method and apparatus for resolving a variable number of potential memory access conflicts in a pipelined computer system |
US5226126A (en) * | 1989-02-24 | 1993-07-06 | Nexgen Microsystems | Processor having plurality of functional units for orderly retiring outstanding operations based upon its associated tags |
US5113514A (en) * | 1989-08-22 | 1992-05-12 | Prime Computer, Inc. | System bus for multiprocessor computer system |
US5185871A (en) | 1989-12-26 | 1993-02-09 | International Business Machines Corporation | Coordination of out-of-sequence fetching between multiple processors using re-execution of instructions |
US5226130A (en) * | 1990-02-26 | 1993-07-06 | Nexgen Microsystems | Method and apparatus for store-into-instruction-stream detection and maintaining branch prediction cache consistency |
EP0447736B1 (en) * | 1990-03-19 | 1995-09-27 | BULL HN INFORMATION SYSTEMS ITALIA S.p.A. | Multiprocessor system having distributed shared resources and dynamic and selective global data replication and method therefor |
DE69130588T2 (en) | 1990-05-29 | 1999-05-27 | Nat Semiconductor Corp | Partially decoded instruction cache and method therefor |
EP0459233A3 (en) | 1990-05-29 | 1992-04-08 | National Semiconductor Corporation | Selectively locking memory locations within a microprocessor's on-chip cache |
US5241663A (en) * | 1990-05-31 | 1993-08-31 | Sony Corporation | Hierarchically pairing memory blocks based upon relative storage capacities and simultaneously accessing each memory block within the paired memory blocks |
EP0461926B1 (en) * | 1990-06-15 | 1998-09-02 | Compaq Computer Corporation | Multilevel inclusion in multilevel cache hierarchies |
US5416907A (en) * | 1990-06-15 | 1995-05-16 | Digital Equipment Corporation | Method and apparatus for transferring data processing data transfer sizes |
US5432918A (en) * | 1990-06-29 | 1995-07-11 | Digital Equipment Corporation | Method and apparatus for ordering read and write operations using conflict bits in a write queue |
IE860318L (en) * | 1990-10-01 | 1986-08-05 | Digital Equipment Corp | System bus for a multi-cache data processing system |
US5963745A (en) | 1990-11-13 | 1999-10-05 | International Business Machines Corporation | APAP I/O programmable router |
US5765011A (en) | 1990-11-13 | 1998-06-09 | International Business Machines Corporation | Parallel processing system having a synchronous SIMD processing with processing elements emulating SIMD operation using individual instruction streams |
US5303362A (en) * | 1991-03-20 | 1994-04-12 | Digital Equipment Corporation | Coupled memory multiprocessor computer system including cache coherency management protocols |
US5325504A (en) * | 1991-08-30 | 1994-06-28 | Compaq Computer Corporation | Method and apparatus for incorporating cache line replacement and cache write policy information into tag directories in a cache system |
GB2263987B (en) | 1992-02-06 | 1996-03-06 | Intel Corp | End bit markers for instruction decode |
GB2263985B (en) | 1992-02-06 | 1995-06-14 | Intel Corp | Two stage window multiplexors for deriving variable length instructions from a stream of instructions |
US5487162A (en) * | 1992-02-25 | 1996-01-23 | Matsushita Electric Industrial Co., Ltd. | Cache lock information feeding system using an address translator |
JPH0619785A (en) * | 1992-03-27 | 1994-01-28 | Matsushita Electric Ind Co Ltd | Distributed shared virtual memory and its constitution method |
EP0568231B1 (en) * | 1992-04-29 | 1999-03-10 | Sun Microsystems, Inc. | Methods and apparatus for providing multiple outstanding operations in a cache consistent multiple processor computer system |
US5353425A (en) * | 1992-04-29 | 1994-10-04 | Sun Microsystems, Inc. | Methods and apparatus for implementing a pseudo-LRU cache memory replacement scheme with a locking feature |
JPH0621346A (en) | 1992-05-05 | 1994-01-28 | Xerox Corp | Integrated linear high-voltage device |
US5809531A (en) * | 1992-09-21 | 1998-09-15 | Intel Corporation | Computer system for executing programs using an internal cache without accessing external RAM |
KR960006484B1 (en) * | 1992-09-24 | 1996-05-16 | 마쯔시다 덴기 산교 가부시끼가이샤 | Cache memory device |
JP3644959B2 (en) * | 1992-09-29 | 2005-05-11 | セイコーエプソン株式会社 | Microprocessor system |
US5638537A (en) | 1993-01-29 | 1997-06-10 | Mitsubishi Denki Kabushiki Kaisha | Cache system with access mode determination for prioritizing accesses to cache memory |
US5493667A (en) * | 1993-02-09 | 1996-02-20 | Intel Corporation | Apparatus and method for an instruction cache locking scheme |
US5410669A (en) * | 1993-04-05 | 1995-04-25 | Motorola, Inc. | Data processor having a cache memory capable of being used as a linear ram bank |
US5450551A (en) * | 1993-05-28 | 1995-09-12 | International Business Machines Corporation | System direct memory access (DMA) support logic for PCI based computer system |
US5416783A (en) * | 1993-08-09 | 1995-05-16 | Motorola, Inc. | Method and apparatus for generating pseudorandom numbers or for performing data compression in a data processor |
IE80854B1 (en) | 1993-08-26 | 1999-04-07 | Intel Corp | Processor ordering consistency for a processor performing out-of-order instruction execution |
US5471598A (en) * | 1993-10-18 | 1995-11-28 | Cyrix Corporation | Data dependency detection and handling in a microprocessor with write buffer |
US5510934A (en) * | 1993-12-15 | 1996-04-23 | Silicon Graphics, Inc. | Memory system including local and global caches for storing floating point and integer data |
US5588126A (en) * | 1993-12-30 | 1996-12-24 | Intel Corporation | Methods and apparatus for fordwarding buffered store data on an out-of-order execution computer system |
US5526510A (en) * | 1994-02-28 | 1996-06-11 | Intel Corporation | Method and apparatus for implementing a single clock cycle line replacement in a data cache unit |
US5671444A (en) * | 1994-02-28 | 1997-09-23 | Intel Corporaiton | Methods and apparatus for caching data in a non-blocking manner using a plurality of fill buffers |
US5634004A (en) | 1994-05-16 | 1997-05-27 | Network Programs, Inc. | Directly programmable distribution element |
JPH07334428A (en) | 1994-06-14 | 1995-12-22 | Toshiba Corp | Cache memory |
US5551001A (en) * | 1994-06-29 | 1996-08-27 | Exponential Technology, Inc. | Master-slave cache system for instruction and data cache memories |
US5644752A (en) * | 1994-06-29 | 1997-07-01 | Exponential Technology, Inc. | Combined store queue for a master-slave cache system |
JPH0816470A (en) * | 1994-07-04 | 1996-01-19 | Hitachi Ltd | Parallel computer |
US5668972A (en) * | 1994-10-05 | 1997-09-16 | International Business Machines Corporation | Method and system for efficient miss sequence cache line allocation utilizing an allocation control cell state to enable a selected match line |
US5592679A (en) * | 1994-11-14 | 1997-01-07 | Sun Microsystems, Inc. | Apparatus and method for distributed control in a processor architecture |
US5586253A (en) * | 1994-12-15 | 1996-12-17 | Stratus Computer | Method and apparatus for validating I/O addresses in a fault-tolerant computer system |
US5584014A (en) * | 1994-12-20 | 1996-12-10 | Sun Microsystems, Inc. | Apparatus and method to preserve data in a set associative memory device |
EP0735487B1 (en) | 1995-03-31 | 2001-10-31 | Sun Microsystems, Inc. | A fast, dual ported cache controller for data processors in a packet switched cache coherent multiprocessor system |
US5761712A (en) * | 1995-06-07 | 1998-06-02 | Advanced Micro Devices | Data memory unit and method for storing data into a lockable cache in one clock cycle by previewing the tag array |
US5539878A (en) * | 1995-06-16 | 1996-07-23 | Elonex Technologies, Inc. | Parallel testing of CPU cache and instruction units |
US5887146A (en) * | 1995-08-14 | 1999-03-23 | Data General Corporation | Symmetric multiprocessing computer with non-uniform memory access architecture |
JPH09101916A (en) * | 1995-10-06 | 1997-04-15 | Fujitsu Ltd | Multiprocess processor |
US5603047A (en) | 1995-10-06 | 1997-02-11 | Lsi Logic Corporation | Superscalar microprocessor architecture |
US5805920A (en) | 1995-11-13 | 1998-09-08 | Tandem Computers Incorporated | Direct bulk data transfers |
US5897651A (en) * | 1995-11-13 | 1999-04-27 | International Business Machines Corporation | Information handling system including a direct access set associative cache and method for accessing same |
US5710907A (en) | 1995-12-22 | 1998-01-20 | Sun Microsystems, Inc. | Hybrid NUMA COMA caching system and methods for selecting between the caching modes |
US6038644A (en) * | 1996-03-19 | 2000-03-14 | Hitachi, Ltd. | Multiprocessor system with partial broadcast capability of a cache coherent processing request |
GB2311880A (en) * | 1996-04-03 | 1997-10-08 | Advanced Risc Mach Ltd | Partitioned cache memory |
JPH09325913A (en) * | 1996-06-05 | 1997-12-16 | Toshiba Corp | Semiconductor memory |
US5887138A (en) | 1996-07-01 | 1999-03-23 | Sun Microsystems, Inc. | Multiprocessing computer system employing local and global address spaces and COMA and NUMA access modes |
US5878268A (en) | 1996-07-01 | 1999-03-02 | Sun Microsystems, Inc. | Multiprocessing system configured to store coherency state within multiple subnodes of a processing node |
US5813029A (en) | 1996-07-09 | 1998-09-22 | Micron Electronics, Inc. | Upgradeable cache circuit using high speed multiplexer |
US5937431A (en) * | 1996-07-12 | 1999-08-10 | Samsung Electronics Co., Ltd. | Multi- node, multi-level cache- only memory architecture with relaxed inclusion |
US5668815A (en) * | 1996-08-14 | 1997-09-16 | Advanced Micro Devices, Inc. | Method for testing integrated memory using an integrated DMA controller |
US5961623A (en) | 1996-08-29 | 1999-10-05 | Apple Computer, Inc. | Method and system for avoiding starvation and deadlocks in a split-response interconnect of a computer system |
US5748640A (en) * | 1996-09-12 | 1998-05-05 | Advanced Micro Devices | Technique for incorporating a built-in self-test (BIST) of a DRAM block with existing functional test vectors for a microprocessor |
US5802338A (en) * | 1996-10-01 | 1998-09-01 | International Business Machines Corporation | Method of self-parallelizing and self-parallelizing multiprocessor using the method |
US5752261A (en) * | 1996-11-07 | 1998-05-12 | Ncr Corporation | Method and apparatus for detecting thrashing in a cache memory |
JPH10154100A (en) | 1996-11-25 | 1998-06-09 | Canon Inc | Information processing system, device and its controlling method |
US5809528A (en) * | 1996-12-24 | 1998-09-15 | International Business Machines Corporation | Method and circuit for a least recently used replacement mechanism and invalidated address handling in a fully associative many-way cache memory |
JP2937919B2 (en) * | 1997-01-16 | 1999-08-23 | 日本電気アイシーマイコンシステム株式会社 | Pseudo random number generator |
US5983321A (en) * | 1997-03-12 | 1999-11-09 | Advanced Micro Devices, Inc. | Cache holding register for receiving instruction packets and for providing the instruction packets to a predecode unit and instruction cache |
US5768555A (en) * | 1997-02-20 | 1998-06-16 | Advanced Micro Devices, Inc. | Reorder buffer employing last in buffer and last in line bits |
US5913224A (en) * | 1997-02-26 | 1999-06-15 | Advanced Micro Devices, Inc. | Programmable cache including a non-lockable data way and a lockable data way configured to lock real-time data |
JP3904282B2 (en) | 1997-03-31 | 2007-04-11 | 株式会社ルネサステクノロジ | Semiconductor integrated circuit device |
US6105119A (en) | 1997-04-04 | 2000-08-15 | Texas Instruments Incorporated | Data transfer circuitry, DSP wrapper circuitry and improved processor devices, methods and systems |
US6298370B1 (en) | 1997-04-04 | 2001-10-02 | Texas Instruments Incorporated | Computer operating process allocating tasks between first and second processors at run time based upon current processor load |
US5898849A (en) * | 1997-04-04 | 1999-04-27 | Advanced Micro Devices, Inc. | Microprocessor employing local caches for functional units to store memory operands used by the functional units |
US5974507A (en) * | 1997-04-14 | 1999-10-26 | International Business Machines Corporation | Optimizing a cache eviction mechanism by selectively introducing different levels of randomness into a replacement algorithm |
US6182201B1 (en) | 1997-04-14 | 2001-01-30 | International Business Machines Corporation | Demand-based issuance of cache operations to a system bus |
FR2762418B1 (en) | 1997-04-17 | 1999-06-11 | Alsthom Cge Alcatel | METHOD FOR MANAGING A SHARED MEMORY |
US5784588A (en) * | 1997-06-20 | 1998-07-21 | Sun Microsystems, Inc. | Dependency checking apparatus employing a scoreboard for a pair of register sets having different precisions |
US6161167A (en) * | 1997-06-27 | 2000-12-12 | Advanced Micro Devices, Inc. | Fully associate cache employing LRU groups for cache replacement and mechanism for selecting an LRU group |
JP3524337B2 (en) | 1997-07-25 | 2004-05-10 | キヤノン株式会社 | Bus management device and control device for multifunction device having the same |
US6295584B1 (en) * | 1997-08-29 | 2001-09-25 | International Business Machines Corporation | Multiprocessor computer system with memory map translation |
ATE247299T1 (en) * | 1997-09-05 | 2003-08-15 | Sun Microsystems Inc | MULTI-PROCESSOR COMPUTER SYSTEM USING A GROUP PROTECTION MECHANISM |
US6185703B1 (en) * | 1997-10-10 | 2001-02-06 | Intel Corporation | Method and apparatus for direct access test of embedded memory |
US6101420A (en) | 1997-10-24 | 2000-08-08 | Compaq Computer Corporation | Method and apparatus for disambiguating change-to-dirty commands in a switch based multi-processing system with coarse directories |
US6108752A (en) | 1997-10-24 | 2000-08-22 | Compaq Computer Corporation | Method and apparatus for delaying victim writes in a switch-based multi-processor system to maintain data coherency |
US6209065B1 (en) | 1997-10-24 | 2001-03-27 | Compaq Computer Corporation | Mechanism for optimizing generation of commit-signals in a distributed shared-memory system |
US6085294A (en) | 1997-10-24 | 2000-07-04 | Compaq Computer Corporation | Distributed data dependency stall mechanism |
US6032228A (en) | 1997-11-26 | 2000-02-29 | International Business Machines Corporation | Flexible cache-coherency mechanism |
US6065077A (en) * | 1997-12-07 | 2000-05-16 | Hotrail, Inc. | Apparatus and method for a cache coherent shared memory multiprocessing system |
US6016533A (en) * | 1997-12-16 | 2000-01-18 | Advanced Micro Devices, Inc. | Way prediction logic for cache array |
US6128706A (en) | 1998-02-03 | 2000-10-03 | Institute For The Development Of Emerging Architectures, L.L.C. | Apparatus and method for a load bias--load with intent to semaphore |
US6141733A (en) | 1998-02-17 | 2000-10-31 | International Business Machines Corporation | Cache coherency protocol with independent implementation of optimized cache operations |
US6295608B1 (en) * | 1998-02-17 | 2001-09-25 | Microsoft Corporation | Optimized allocation of data elements among cache lines |
US6289419B1 (en) | 1998-03-06 | 2001-09-11 | Sharp Kabushiki Kaisha | Consistency control device merging updated memory blocks |
US6631448B2 (en) * | 1998-03-12 | 2003-10-07 | Fujitsu Limited | Cache coherence unit for interconnecting multiprocessor nodes having pipelined snoopy protocol |
US6070215A (en) | 1998-03-13 | 2000-05-30 | Compaq Computer Corporation | Computer system with improved transition to low power operation |
GB9806184D0 (en) | 1998-03-23 | 1998-05-20 | Sgs Thomson Microelectronics | A cache coherency mechanism |
GB2335762C (en) * | 1998-03-25 | 2008-01-10 | Advanced Risc Mach Ltd | Write buffering in a data processing apparatus |
GB2335764B (en) * | 1998-03-27 | 2002-10-09 | Motorola Ltd | Circuit and method of controlling cache memory |
US6240532B1 (en) * | 1998-04-06 | 2001-05-29 | Rise Technology Company | Programmable hit and write policy for cache memory test |
US6185657B1 (en) * | 1998-04-20 | 2001-02-06 | Motorola Inc. | Multi-way cache apparatus and method |
US6003106A (en) * | 1998-05-27 | 1999-12-14 | International Business Machines Corporation | DMA cache control logic |
US6351789B1 (en) * | 1998-05-29 | 2002-02-26 | Via-Cyrix, Inc. | Built-in self-test circuit and method for validating an associative data array |
US6195739B1 (en) | 1998-06-29 | 2001-02-27 | Cisco Technology, Inc. | Method and apparatus for passing data among processor complex stages of a pipelined processing engine |
US6266731B1 (en) | 1998-09-03 | 2001-07-24 | Compaq Computer Corporation | High speed peripheral interconnect apparatus, method and system |
JP2000200221A (en) | 1998-10-30 | 2000-07-18 | Nec Corp | Cache memory device and its control method |
US6272522B1 (en) | 1998-11-17 | 2001-08-07 | Sun Microsystems, Incorporated | Computer data packet switching and load balancing system using a general-purpose multiprocessor architecture |
US6338122B1 (en) | 1998-12-15 | 2002-01-08 | International Business Machines Corporation | Non-uniform memory access (NUMA) data processing system that speculatively forwards a read request to a remote processing node |
US6631401B1 (en) | 1998-12-21 | 2003-10-07 | Advanced Micro Devices, Inc. | Flexible probe/probe response routing for maintaining coherency |
US6240432B1 (en) * | 1998-12-28 | 2001-05-29 | Vanguard International Semiconductor Corporation | Enhanced random number generator |
US6425060B1 (en) | 1999-01-05 | 2002-07-23 | International Business Machines Corporation | Circuit arrangement and method with state-based transaction scheduling |
JP2000222280A (en) * | 1999-01-19 | 2000-08-11 | Texas Instr Inc <Ti> | Post-write buffer for double clock system |
US6192452B1 (en) * | 1999-02-26 | 2001-02-20 | International Business Machines Corporation | Method and system for avoiding data loss due to cancelled transactions within a non-uniform memory access system |
US6480489B1 (en) | 1999-03-01 | 2002-11-12 | Sun Microsystems, Inc. | Method and apparatus for data re-assembly with a high performance network interface |
US6249843B1 (en) | 1999-08-05 | 2001-06-19 | International Business Machines Corporation | Store instruction having horizontal memory hierarchy control bits |
US6519690B1 (en) * | 1999-08-23 | 2003-02-11 | Advanced Micro Devices, Inc. | Flexible address programming with wrap blocking |
US6405287B1 (en) * | 1999-11-17 | 2002-06-11 | Hewlett-Packard Company | Cache line replacement using cache status to bias way selection |
US6430655B1 (en) * | 2000-01-31 | 2002-08-06 | Mips Technologies, Inc. | Scratchpad RAM memory accessible in parallel to a primary cache |
US20010052053A1 (en) | 2000-02-08 | 2001-12-13 | Mario Nemirovsky | Stream processing unit for a multi-streaming processor |
US6460124B1 (en) * | 2000-10-20 | 2002-10-01 | Wisconsin Alumni Research Foundation | Method of using delays to speed processing of inferred critical program portions |
US6748501B2 (en) * | 2000-12-30 | 2004-06-08 | International Business Machines Corporation | Microprocessor reservation mechanism for a hashed address system |
US6801986B2 (en) * | 2001-08-20 | 2004-10-05 | Hewlett-Packard Development Company, L.P. | Livelock prevention by delaying surrender of ownership upon intervening ownership request during load locked / store conditional atomic memory operation |
US6799236B1 (en) * | 2001-11-20 | 2004-09-28 | Sun Microsystems, Inc. | Methods and apparatus for executing code while avoiding interference |
EP1363193B1 (en) | 2002-05-15 | 2006-05-03 | Broadcom Corporation | Programmable cache for the partitioning of local and remote cache blocks |
US7266587B2 (en) * | 2002-05-15 | 2007-09-04 | Broadcom Corporation | System having interfaces, switch, and memory bridge for CC-NUMA operation |
-
2002
- 2002-10-11 US US10/270,028 patent/US7266587B2/en not_active Expired - Fee Related
- 2002-11-20 DE DE60207210T patent/DE60207210T2/en not_active Expired - Lifetime
- 2002-11-20 EP EP02025689A patent/EP1363196B1/en not_active Expired - Lifetime
- 2002-11-20 AT AT02025689T patent/ATE309574T1/en not_active IP Right Cessation
-
2003
- 2003-04-15 US US10/413,916 patent/US20030229676A1/en not_active Abandoned
- 2003-04-15 US US10/414,363 patent/US6948035B2/en not_active Expired - Lifetime
- 2003-04-15 US US10/413,917 patent/US6988168B2/en not_active Expired - Fee Related
- 2003-05-09 US US10/435,189 patent/US7343456B2/en not_active Expired - Fee Related
- 2003-05-15 US US10/439,343 patent/US7340546B2/en not_active Expired - Fee Related
-
2007
- 2007-08-14 US US11/891,954 patent/US7469275B2/en not_active Expired - Fee Related
Patent Citations (41)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4463424A (en) * | 1981-02-19 | 1984-07-31 | International Business Machines Corporation | Method for dynamically allocating LRU/MRU managed memory among concurrent sequential processes |
US4433378A (en) * | 1981-09-28 | 1984-02-21 | Western Digital | Chip topography for MOS packet network interface circuit |
US4760571A (en) * | 1984-07-25 | 1988-07-26 | Siegfried Schwarz | Ring network for communication between one chip processors |
US5974508A (en) * | 1992-07-31 | 1999-10-26 | Fujitsu Limited | Cache memory system and method for automatically locking cache entries to prevent selected memory items from being replaced |
US5914955A (en) * | 1993-10-20 | 1999-06-22 | Lsi Logic Corporation | Switched network hub on a chip |
US5640399A (en) * | 1993-10-20 | 1997-06-17 | Lsi Logic Corporation | Single chip network router |
US5668809A (en) * | 1993-10-20 | 1997-09-16 | Lsi Logic Corporation | Single chip network hub with dynamic window filter |
US5802287A (en) * | 1993-10-20 | 1998-09-01 | Lsi Logic Corporation | Single chip universal protocol multi-function ATM network interface |
US5887187A (en) * | 1993-10-20 | 1999-03-23 | Lsi Logic Corporation | Single chip network adapter apparatus |
US6021451A (en) * | 1994-05-20 | 2000-02-01 | Intel Corporation | Method and apparatus for maintaining transaction ordering and arbitrating in a bus bridge |
US5778438A (en) * | 1995-12-06 | 1998-07-07 | Intel Corporation | Method and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests |
US6373846B1 (en) * | 1996-03-07 | 2002-04-16 | Lsi Logic Corporation | Single chip networking device with enhanced memory access co-processor |
US5778414A (en) * | 1996-06-13 | 1998-07-07 | Racal-Datacom, Inc. | Performance enhancing memory interleaver for data frame processing |
US5893150A (en) * | 1996-07-01 | 1999-04-06 | Sun Microsystems, Inc. | Efficient allocation of cache memory space in a computer system |
US6366583B2 (en) * | 1996-08-07 | 2002-04-02 | Cisco Technology, Inc. | Network router integrated onto a silicon chip |
US6209020B1 (en) * | 1996-09-20 | 2001-03-27 | Nortel Networks Limited | Distributed pipeline memory architecture for a computer system with even and odd pids |
US6202125B1 (en) * | 1996-11-25 | 2001-03-13 | Intel Corporation | Processor-cache protocol using simple commands to implement a range of cache configurations |
US5829025A (en) * | 1996-12-17 | 1998-10-27 | Intel Corporation | Computer system and method of allocating cache memories in a multilevel cache hierarchy utilizing a locality hint within an instruction |
US6266797B1 (en) * | 1997-01-16 | 2001-07-24 | Advanced Micro Devices, Inc. | Data transfer network on a computer chip using a re-configurable path multiple ring topology |
US6111859A (en) * | 1997-01-16 | 2000-08-29 | Advanced Micro Devices, Inc. | Data transfer network on a computer chip utilizing combined bus and ring topologies |
US6157623A (en) * | 1997-02-14 | 2000-12-05 | Advanced Micro Devices, Inc. | Apparatus and method for selectively outputting data using a MAC layer interface or a PCI bus interface |
US6018763A (en) * | 1997-05-28 | 2000-01-25 | 3Com Corporation | High performance shared memory for a bridge router supporting cache coherency |
US6128677A (en) * | 1997-10-15 | 2000-10-03 | Intel Corporation | System and method for improved transfer of data between multiple processors and I/O bridges |
US5908468A (en) * | 1997-10-24 | 1999-06-01 | Advanced Micro Devices, Inc. | Data transfer network on a chip utilizing a multiple traffic circle topology |
US6092137A (en) * | 1997-11-26 | 2000-07-18 | Industrial Technology Research Institute | Fair data bus arbitration system which assigns adjustable priority values to competing sources |
US6151662A (en) * | 1997-12-02 | 2000-11-21 | Advanced Micro Devices, Inc. | Data transaction typing for improved caching and prefetching characteristics |
US6279087B1 (en) * | 1997-12-22 | 2001-08-21 | Compaq Computer Corporation | System and method for maintaining coherency and improving performance in a bus bridge supporting write posting operations |
US6202129B1 (en) * | 1998-03-31 | 2001-03-13 | Intel Corporation | Shared cache structure for temporal and non-temporal information using indicative bits |
US6321309B1 (en) * | 1998-05-19 | 2001-11-20 | Lsi Logic Corporation | Memory arbitration scheme with circular sequence register |
US6098064A (en) * | 1998-05-22 | 2000-08-01 | Xerox Corporation | Prefetching and caching documents according to probability ranked need S list |
US6215497B1 (en) * | 1998-08-12 | 2001-04-10 | Monolithic System Technology, Inc. | Method and apparatus for maximizing the random access bandwidth of a multi-bank DRAM in a computer graphics system |
US6378048B1 (en) * | 1998-11-12 | 2002-04-23 | Intel Corporation | “SLIME” cache coherency system for agents with multi-layer caches |
US6266743B1 (en) * | 1999-02-26 | 2001-07-24 | International Business Machines Corporation | Method and system for providing an eviction protocol within a non-uniform memory access system |
US6269427B1 (en) * | 1999-03-18 | 2001-07-31 | International Business Machines Corporation | Multiple load miss handling in a cache memory system |
US6332179B1 (en) * | 1999-08-19 | 2001-12-18 | International Business Machines Corporation | Allocation for back-to-back misses in a directory based cache |
US6349365B1 (en) * | 1999-10-08 | 2002-02-19 | Advanced Micro Devices, Inc. | User-prioritized cache replacement |
US6438651B1 (en) * | 1999-11-01 | 2002-08-20 | International Business Machines Corporation | Method, system, and program for managing requests to a cache using flags to queue and dequeue data in a buffer |
US6262594B1 (en) * | 1999-11-05 | 2001-07-17 | Ati International, Srl | Apparatus and method for configurable use of groups of pads of a system on chip |
US6519685B1 (en) * | 1999-12-22 | 2003-02-11 | Intel Corporation | Cache states for multiprocessor cache coherency protocols |
US6681293B1 (en) * | 2000-08-25 | 2004-01-20 | Silicon Graphics, Inc. | Method and cache-coherence system allowing purging of mid-level cache entries without purging lower-level cache entries |
US6574708B2 (en) * | 2001-05-18 | 2003-06-03 | Broadcom Corporation | Source controlled cache allocation |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040215473A1 (en) * | 2003-04-24 | 2004-10-28 | Sun Microsystems, Inc. | Simultaneous global transaction and local transaction management in an application server |
US7610305B2 (en) | 2003-04-24 | 2009-10-27 | Sun Microsystems, Inc. | Simultaneous global transaction and local transaction management in an application server |
US20050015353A1 (en) * | 2003-07-14 | 2005-01-20 | Sun Microsystems, Inc. | Read/write lock transaction manager freezing |
US7739252B2 (en) * | 2003-07-14 | 2010-06-15 | Oracle America, Inc. | Read/write lock transaction manager freezing |
US20060003439A1 (en) * | 2004-07-02 | 2006-01-05 | Ismagilov Rustem F | Microfluidic system |
US20060143406A1 (en) * | 2004-12-27 | 2006-06-29 | Chrysos George Z | Predictive early write-back of owned cache blocks in a shared memory computer system |
US7624236B2 (en) * | 2004-12-27 | 2009-11-24 | Intel Corporation | Predictive early write-back of owned cache blocks in a shared memory computer system |
Also Published As
Publication number | Publication date |
---|---|
US7340546B2 (en) | 2008-03-04 |
US20070282968A1 (en) | 2007-12-06 |
DE60207210T2 (en) | 2006-07-27 |
US20030217115A1 (en) | 2003-11-20 |
DE60207210D1 (en) | 2005-12-15 |
US6988168B2 (en) | 2006-01-17 |
US20030217216A1 (en) | 2003-11-20 |
US20030233495A1 (en) | 2003-12-18 |
US7343456B2 (en) | 2008-03-11 |
EP1363196A1 (en) | 2003-11-19 |
US20030217238A1 (en) | 2003-11-20 |
US6948035B2 (en) | 2005-09-20 |
EP1363196B1 (en) | 2005-11-09 |
US20030217229A1 (en) | 2003-11-20 |
ATE309574T1 (en) | 2005-11-15 |
US7266587B2 (en) | 2007-09-04 |
US7469275B2 (en) | 2008-12-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20030229676A1 (en) | Command to transfer data from node state agent to memory bridge | |
US7269695B2 (en) | Ambiguous virtual channels | |
US7003631B2 (en) | System having address-based intranode coherency and data-based internode coherency | |
US6993631B2 (en) | L2 cache maintaining local ownership of remote coherency blocks | |
US6748479B2 (en) | System having interfaces and switch that separates coherent and packet traffic | |
US7424561B2 (en) | Systems using mix of packet, coherent, and noncoherent traffic to optimize transmission between systems | |
US6571321B2 (en) | Read exclusive for fast, simple invalidate | |
US6965973B2 (en) | Remote line directory which covers subset of shareable CC-NUMA memory space | |
EP1363193B1 (en) | Programmable cache for the partitioning of local and remote cache blocks | |
EP1363188B1 (en) | Load-linked/store conditional mechanism in a cc-numa (cache-coherent nonuniform memory access) system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: BROADCOM CORPORATION, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:ROWLANDS, JOSEPH B.;REEL/FRAME:013978/0661 Effective date: 20030409 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |
|
AS | Assignment |
Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH CAROLINA Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001 Effective date: 20160201 Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001 Effective date: 20160201 |
|
AS | Assignment |
Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD., SINGAPORE Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001 Effective date: 20170120 Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001 Effective date: 20170120 |
|
AS | Assignment |
Owner name: BROADCOM CORPORATION, CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENTS;ASSIGNOR:BANK OF AMERICA, N.A., AS COLLATERAL AGENT;REEL/FRAME:041712/0001 Effective date: 20170119 |