摘要:
An error correction apparatus includes an error control circuit which computes for each burst of a message (for a destination unit) an error correction code as a function of an initial error correction code at the first burst of the message or of the error correction code of the previous burst and of the data bytes of the burst. The burst error correction code is sent on a medium which is separate from the data transport medium as a companion of the burst. Also, the error control circuit receives the burst error correction code from an origin unit and generates the burst error correction code to be compared with the received burst error correction code. If a mismatch is detected, the burst found in error is flagged.
摘要:
The system performs an optimized number of simultaneous transfers of data packets between pairs of units comprising an origin unit and a target unit selected among N data processing units (8). Each data processing unit comprises a set of outbound queues with one outbound queue associated with each one of the data processing units to which it may send data packets, for storing the data packets to be sent by the data processing unit to the data processing unit associated with said one outbound queue. The transfers are performed during a time burst Ti+1 by data switch 6 under control of switching control signals sent to data switch by the units on lines 16-1 to 16-N in response to control out signals generated by scheduler 4 during previous burst time Ti. The scheduler runs a selection algorithm which gives each unit an equal probability to be selected as origin unit in a given period.
摘要:
The synchronization circuit resynchronizes the data bits received from remote devices on line or link (20-1) with their own clock CS and frame synchronization signal FS with a central clock CO and central frame synchronization signal FO. The received bits are sequentially arranged in an n-bit cyclic buffer (114-1) with the received bit clock CS. The arranged bits are sequentially picked at the opposite buffer position with the central clock CO. The buffer loading position is provided by binary counter 102 incremented by CS and the buffer picking portion is given by binary counter 100 incremented by CO. At initialization counters 102 and 100 are set to 0 and n/2. The resynchronized data bits on line 21-1 and the resynchronized frame signal FSR on line 61-10 are provided to an additional circuit which synchronize the data bits at the frame level.
摘要:
A method of operating a packet parser in a computing system includes providing a configurable packet pointer by the packet parser, the packet pointer configured to index a configurable number of atomic parsing elements, the atomic parsing elements having a configurable size, in a data stream received by the computing system for extraction, wherein the indexed atomic parsing elements are non-contiguous in the data stream; and receiving the extracted indexed atomic parsing elements from the data stream by the packet parser.
摘要:
A system and method in accordance with the present invention allows for an adapter to be utilized in a server environment that can accommodate both a 10 G and a 1 G source utilizing the same pins. This is accomplished through the use of a high speed serializer/deserializer (high speed serdes) which can accommodate both data sources. The high speed serdes allows for the use of a relatively low reference clock speed on the NIC to provide the proper clocking of the data sources and also allows for different modes to be set to accommodate the different data sources. Finally the system allows for the adapter to use the same pins for multiple data sources.
摘要:
A system and method for computing a blind checksum includes a host Ethernet adapter (HEA) with a system for receiving a packet. The system determines whether or not the packet is in Internet protocol version four (IPv4). If the packet is not in IPv4, the system computes the checksum of the packet. If the packet is in IPv4, the system determines whether the packet is in transmission control protocol (TCP) or user datagram protocol (UDP). If the packet is not in either of TCP or UDP the system attaches a pseudo-header to the packet and computes the checksum of the packet based on the pseudo-header and the IPv4 standard.
摘要:
Providing communications between operating system partitions and a computer network. In one aspect, an apparatus for distributing network communications among multiple operating system partitions includes a physical port allowing communications between the network and the computer system, and logical ports associated with the physical port, where each logical port is associated with one of the operating system partitions. Each of the logical ports enables communication between a physical port and the associated operating system partition and allows configurability of network resources of the system. Other aspects include a logical switch for logical and physical ports, and packet queues for each connection and for each logical port.
摘要:
A system and method for parsing, filtering, and computing the checksum in a host Ethernet adapter (HEA) that is coupled to a host. The method includes receiving a part of a frame, wherein a plurality of parts of a frame constitute a entire frame. Next, parse the part of a frame before receiving the entire frame. The HEA computes a checksum of the part of a frame. The HEA filters the part of a frame based on a logical, port-specific policy and transmits the checksum to the host.
摘要:
Systems and methods for implementing multi-frame control blocks in a network processor are disclosed. Embodiments include systems and methods to reduce long latency memory access to less expensive memory such as DRAM. As a network processor in a network receives packets of data, the network processor forms a frame control block for each packet. The frame control block contains a pointer to a memory location where the packet data is stored, and is thereby associated with the packet. The network processor associates a plurality of frame control blocks together in a table control block that is stored in a control store. Each table control block comprises a pointer to a memory location of a next table control block in a chain of table control blocks. Because frame control blocks are stored and accessed in table control blocks, less frequent memory accesses may be needed to keep up with the frame rate of packet transmission.
摘要:
Systems and methods for adaptively mapping system memory address bits into an instruction tag and an index into the cache are disclosed. More particularly, hardware and software are disclosed for observing collisions that occur for a given mapping of system memory bits into a tag and an index. Based on the observations, an optimal mapping may be determined that minimizes collisions.