Abstract: A system and method are disclosed for connecting PCI-ordered agents based on fully independent networks. The system and method are free of PCI topology constraints, so that the system and method can be implemented in an inexpensive and scalable way. The method disclosed is used to handle and transport PCI-ordered traffic on a fabric. Based on the actual ordering requirement of the set of PCI agents, the fabric includes two, three, or four independent networks.

Abstract: A multiple processor device generates a control packet for at least one connectionless-based packet in partial accordance with a control packet format of the connection-based point-to-point link and partially not in accordance with the control packet format. For instance, the multiple processor device generates the control packet to include, in noncompliance with the control packet format, one or more of an indication that at least one connectionless-based packet is being transported, an indication of a virtual channel of a plurality of virtual channels associated with the at least one connectionless-based packet, an indication of an amount of data included in the associated data packet, status of the at least one connectionless-based packet, and an error status indication.

Abstract: The present disclosure provides methods and systems adapted for use with a processor having one or more physical cores. The methods and systems include a virtual core management component adapted to map one or more virtual cores to at least one of the physical cores to enable execution of one or more programs by the at least one physical core. The one or more virtual cores include one or more logical states associated with the execution of the one or more programs. The methods and systems may include a memory component adapted to store the one or more virtual cores. The virtual core management component may be adapted to transfer the one or more virtual cores from the memory component to the at least one physical core.

Abstract: A multiple processor device generates a control packet for at least one connectionless-based packet in partial accordance with a control packet format of the connection-based point-to-point link and partially not in accordance with the control packet format. For instance, the multiple processor device generates the control packet to include, in noncompliance with the control packet format, one or more of an indication that at least one connectionless-based packet is being transported, an indication of a virtual channel of a plurality of virtual channels associated with the at least one connectionless-based packet, an indication of an amount of data included in the associated data packet, status of the at least one connectionless-based packet, and an error status indication.

Abstract: A multiple processor device generates a control packet for at least one connectionless-based packet in partial accordance with a control packet format of the connection-based point-to-point link and partially not in accordance with the control packet format. For instance, the multiple processor device generates the control packet to include, in noncompliance with the control packet format, one or more of an indication that at least one connectionless-based packet is being transported, an indication of a virtual channel of a plurality of virtual channels associated with the at least one connectionless-based packet, an indication of an amount of data included in the associated data packet, status of the at least one connectionless-based packet, and an error status indication.

Abstract: In accordance with the present invention a system for detecting transaction errors in a system comprising a plurality of data processing devices using a common system interconnect bus, comprises a node controller operably connected to said system interconnect bus and a plurality of interface agents communicatively coupled to said node controller. Error corresponding to transactions between said interface agents and other processing modules in said system are directed to said node controller; and wherein transaction errors that would not normally be communicated to said system interconnect bus are communicated by said node controller to said system interconnect bus to be available for detection. In an embodiment of the present invention, the interface agents operate in accordance with the hypertransport protocol. A system control and debug unit and a trace cache operably connected to the system bus can be used to diagnose and store errors conditions.

Abstract: A multiprocessor switching device substantially implemented on a single CMOS integrated circuit is described in connection with a parallel routing scheme for calculating routing information for incoming packets. Using the programmable hash and route routing scheme, a hash and route circuit can be programmed for a variety of applications, such as routing, flow-splitting or load balancing.

Abstract: According to the present invention, the multiple processor device determines routing for a plurality of data segments. In determining the routing, the multiple processor device first receives the plurality of data segments. The plurality of data segments include multiplexed data fragments from at least one of a plurality of virtual channels. Further, a data segment of the plurality of data segments corresponds to one of the multiplexed data fragments. The multiple processor device then applies at least one routing rule to one of the plurality of data segments to produce at least one result corresponding to the one of the plurality of data segments. The multiple processor device then interprets the at least one result to determine whether sufficient information is available to render a routing decision for the one of the plurality of data segments.

Abstract: A system and method are provided for increasing the number of processors on a single integrated circuit to a number that is larger than would typically be possible to coordinate on a single bus. In an embodiment of the present invention a two-level memory coherency scheme is implemented for use by multiple processors operably coupled to multiple buses in the same integrated circuit. A control device, such as node controller, is used to control traffic between the two coherency levels. In an embodiment of the invention the first level of coherency is implemented using a “snoopy” protocol and the second level of coherency is a directory-based coherency scheme.

Abstract: A multiprocessor switching device substantially implemented on a single CMOS integrated circuit is described in connection with a parallel routing scheme for calculating routing information for incoming packets. Using the programmable hash and route routing scheme, a hash and route circuit can be programmed for a variety of applications, such as routing, flow-splitting or load balancing.

Abstract: A multiple processor device stores a stream of data as a plurality of data segments, which includes multiplexed data fragments from at least one of a plurality of virtual channels. The data segments that comprise the stream of data correspond to the multiplexed data fragments from the virtual channels. The multiple processor device then decodes at least one data segment in accordance with one of a plurality of transmission protocols to produce a decoded data segment. The multiple processor device then stores the decoded data segment to align it in accordance with a data path segment size. The multiple processor device then interprets the stored decoded data segment with respect to a corresponding one of the plurality of virtual channels to determine a destination of the stored decoded data segment. The multiple processor device then stores the decoded data segment as part of reassembled data.

Abstract: An integrated circuit includes receive circuits for receiving packets, transmit circuits for transmitting packets, a packet DMA circuit for communicating packets to and from a memory controller, and a switch for selectively coupling the receive circuits to transmit circuits. The integrated circuit may flexibly merge and split the packet streams to provide for various packet processing/packet routing functions to be applied to different packets within the packet streams. An apparatus may include two or more of the integrated circuits, which may communicate packets between respective receive and transmit circuits.

Abstract: A system for servicing data transactions within a processing device using common data paths. The system is broadly comprised of: a plurality of source agents operable to transmit a plurality of data cells; a plurality of destination agents operable to of data cells; a plurality of virtual channels for transporting data cells between the source agents and the destination agents; and a switch. The switch is operable to connect predetermined combinations of the source agents and the destination agents for the transmission of data. The switch generates a plurality of switch processing cycles and processes a plurality of control signals during the switch processing cycles.

Abstract: Power conservation via DRAM access reduction is provided by a buffer/mini-cache selectively operable in a normal mode and a buffer mode. In the buffer mode, entered when CPUs begin operating in low-power states, non-cacheable accesses (such as generated by a DMA device) matching specified physical address ranges are processed by the buffer/mini-cache, instead of by a memory controller and DRAM. The buffer/mini-cache processing includes allocating lines when references miss, and returning cached data from the buffer/mini-cache when references hit. Lines are replaced in the buffer/mini-cache according to one of a plurality of replacement policies, including ceasing replacement when there are no available free lines. In the normal mode, entered when CPUs begin operating in high-power states, the buffer/mini-cache operates akin to a conventional cache and non-cacheable accesses are not processed therein.

Abstract: An apparatus may include a first system and a second system. The first system includes a first plurality of interface circuits, and each of the first plurality of interface circuits is configured to couple to a separate interface. The second system includes a second plurality of interface circuits, and each of the second plurality of interface circuits is configured to couple to a separate interface. A first interface circuit of the first plurality of interface circuits and a second interface circuit of the second plurality of interface circuits are coupled to a first interface. Both the first interface circuit and the second interface circuit are configured to communicate packets, coherency commands, and noncoherent commands on the first interface.

Abstract: A method for receiving data from a plurality of virtual channels begins by storing a stream of data as a plurality of data segments, wherein the stream of data includes multiplexed data fragments from at least one of the plurality of virtual channels, and wherein a data segment of the plurality of data segments corresponds to one of the multiplexed data fragments. The method continues by decoding at least one of the plurality of data segments in accordance with one of a plurality of data transmission protocols to produce at least one decoded data segment. The method continues by storing the at least one decoded data segment, in a generic format, to reassemble at least a portion of a packet provided by the at least one of the plurality of virtual channels. The method continues by routing the at least one decoded data segment as at least part of the reassembled packet to one of a plurality of destinations in accordance with the at least one of the plurality of virtual channels.

Abstract: Small and power-efficient buffer/mini-cache sources and sinks selected DMA accesses directed to a memory space included in a coherency domain of a microprocessor when cached data in the microprocessor is inaccessible due to any or all of the microprocessor being in a low-power state not supporting snooping. Satisfying the selected DMA accesses via the buffer/mini-cache enables reduced power consumption by allowing the microprocessor (or portion thereof) to remain in the low-power state. The buffer/mini-cache may be operated (temporarily) incoherently with respect to the cached data in the microprocessor and flushed before deactivation to synchronize with the cached data when the microprocessor (or portion thereof) transitions to a high-power state that enables snooping. Alternatively the buffer/mini-cache may be operated in a manner (incrementally) coherent with the cached data.

Abstract: A multiprocessor switching device substantially implemented on a single CMOS integrated circuit is described in connection with a packet data transfer circuit that uses a fragment storage buffer to align and/or merge data being transferred to or from memory on a plurality of channels. In a packet reception embodiment, a data shifter and fragment store buffer are used to align received packet data to any required offset. The aligned data may and then be written to the system bus or combined with data fragments from prior data cycles before being written to the system bus. When packet data is being transferred to memory on a plurality of channels, the fragment storage may be channelized using register files or flip-flops to store intermediate values of packets and states for each channel.

Abstract: A small and power-efficient buffer/mini-cache sources and sinks selected DMA accesses directed to a memory space included in a coherency domain of a microprocessor when cached data in the microprocessor is inaccessible due to any or all of the microprocessor being in a low-power state not supporting snooping. Satisfying the selected DMA accesses via the buffer/mini-cache enables reduced power consumption by allowing the microprocessor (or portion thereof) to remain in the low-power state. The buffer/mini-cache may be operated (temporarily) incoherently with respect to the cached data in the microprocessor and flushed before deactivation to synchronize with the cached data when the microprocessor (or portion thereof) transitions to a high-power state that enables snooping. Alternatively the buffer/mini-cache may be operated in a manner (incrementally) coherent with the cached data.

Abstract: Power conservation via DRAM access reduction is provided by a buffer/mini-cache selectively operable in a normal mode and a buffer mode. In the buffer mode, entered when CPUs begin operating in low-power states, non-cacheable accesses (such as generated by a DMA device) matching specified physical address ranges are processed by the buffer/mini-cache, instead of by a memory controller and DRAM. The buffer/mini-cache processing includes allocating lines when references miss, and returning cached data from the buffer/mini-cache when references hit. Lines are replaced in the buffer/mini-cache according to one of a plurality of replacement policies, including ceasing replacement when there are no available free lines. In the normal mode, entered when CPUs begin operating in high-power states, the buffer/mini-cache operates akin to a conventional cache and non-cacheable accesses are not processed therein.