Abstract: A method may involve: receiving an even number of odd-sized packets for transmission over a double data rate link; re-packetizing the even number of odd-sized packets into several even-sized packets; transmitting the even-sized packets over the double data rate link; and de-packetizing the even-sized packets to reform the even number of odd-sized packets. Re-packetizing may involve dividing each of the even number of odd-sized packets into an even-sized portion and a remaining portion. Each even-sized portion may be transferred as one of the even-sized packets. The remaining portions of each of the even number of odd-sized packets may be combined to form another one of even-sized packets. De-packetizing may involve associating each of several portions of one of the even-sized packets with a respective other one of the even-sized packets.

Abstract: A method may involve: receiving a request to perform a block write to a target device and data associated with the block write; buffering the data associated with the block write prior to completing the block write to the target device; storing an indication identifying a block address range of the block write; receiving a request to perform a read to the target device; and determining whether an address of the read is within the block address range of the block write identified by the indication. In some embodiments, such a method may allow the read to complete ahead of the block write dependent on whether the address of the read is within the block address range of the block write identified by the indication.

Abstract: A method may involve: receiving a request to perform a block write to a target device and data associated with the block write; buffering the data associated with the block write prior to completing the block write to the target device; storing an indication identifying a block address range of the block write; receiving a request to perform a read to the target device; and determining whether an address of the read is within the block address range of the block write identified by the indication. In some embodiments, such a method may allow the read to complete ahead of the block write dependent on whether the address of the read is within the block address range of the block write identified by the indication.

Abstract: A method may involve: receiving an even number of odd-sized packets for transmission over a double data rate link; re-packetizing the even number of odd-sized packets into several even-sized packets; transmitting the even-sized packets over the double data rate link; and de-packetizing the even-sized packets to reform the even number of odd-sized packets. Re-packetizing may involve dividing each of the even number of odd-sized packets into an even-sized portion and a remaining portion. Each even-sized portion may be transferred as one of the even-sized packets. The remaining portions of each of the even number of odd-sized packets may be combined to form another one of even-sized packets. De-packetizing may involve associating each of several portions of one of the even-sized packets with a respective other one of the even-sized packets.

Abstract: An apparatus includes a memory, a functional unit configured to perform a block operation on one or more block operands to generate a block result, and a cache accumulator that includes multiple block storage locations. The cache accumulator is configured to receive a set of one or more instructions to perform a first accumulation operation. In response to receiving the first instruction in the set, the cache accumulator is configured to access an associativity list that includes an indication that a first set of the block storage locations are currently allocated to the first accumulation operation. In response to the indication, the cache accumulator is configured to provide the first block operand to the functional unit from the first set of the block storage locations and to store the block result into the first set of block storage locations.

Abstract: A real time data processing system consisting of a plurality of processing nodes and a write only reflective data link for transferring information containing writes only between the plurality or processing nodes. All the nodes include a bus, a processor coupled to the bus, a memory having at least two ports with one port connected to the bus and the other port connected to the data link and a sensor for sensing a write to the memory. At least one node has a VMEbus as the bus and serves as an I/O connected to one port of the memory. Further a local bus included for inputting and outputting from the memory. The local bus is connected to a third port of the memory.