| 摘要 |
Methods, systems, and computer program products for high-performance interprocess communication. Each process dynamically identifies routines responsible for managing communication received from other processes through a shared memory heap and a shared memory queue, each of the routines handling one or more operation codes. An allocation from the shared heap produces a process agnostic memory handle from which a process specific memory pointer may be obtained. Using the memory pointer, the enqueuing process places an operation code, parameters, and any other relevant data in the allocated memory and adds the memory handle to a shared queue. The dequeuing process removes the memory handle from the queue and generates a memory pointer to access the allocated memory in the dequeuing process. Upon retrieving the operation code from the allocated memory, the dequeuing process calls the appropriate handler routine. Enqueues may be registered to account for expected responses that are not received.
|
