Multiprocessing system configured to perform efficient block copy operations

An apparatus for performing block copy operations, comprising:

• a processor including a memory management unit configured to translate a virtual address of a memory operation to a local physical address, wherein a block copy operation is specified if said local physical address resides in a specific predefined address space;
• and
• a system interface coupled to receive said block copy operation from said processor, wherein said system interface is configured to perform a translation from said local physical address to a global address and is further configured to transmit a read request including said global address via a network in response to said block copy operation, and wherein said system interface includes a translation storage for storing information for performing said translation from said local physical address to said global address.

An apparatus configured to perform efficient block copy operations, comprising:

• a processor configured to initiate a block copy write to at least one coherency unit within a destination block, wherein said destination block is located within a local processing node which includes said processor;
• and
• a system interface configured to detect said block copy write within said local processing node and to transmit a read request for a corresponding coherency unit within a source block located within a remote processing node, and wherein said system interface transmits said read request upon detection of said block copy write, and wherein said system interface is further configured to receive data from said corresponding coherency unit of said source block and to store said data into said coherency unit within said destination block.

An apparatus for performing block copies comprising:

• a processor configured to execute a block copy command identifying a first coherency unit within a source block and a second coherency unit within a destination block;
• and
• a system interface coupled to receive said block copy command, wherein said system interface is configured to transfer data from said first coherency unit to said second coherency unit in response to said block copy command.