摘要:
The present invention provides for a computer system having a plurality of parallel processor units with each processor unit associated with at least one register for receiving data for the processor unit. The computer system has a bus unit, coupled to the output of each processor unit and the associated register of each processor unit, to transfer the output data of a first processor unit into an associated register of a second processor unit in a single computer operation. The second processor unit is prevented from reading the associated register until the bus unit transfers the output data from the first processor unit to the second processor unit.
摘要:
A computer system simulator concurrently models both processor operation and signal logic behavior and provides a high degree of user interaction and flexibility in the observation and control of signal values and memory contents during the execution of a simulation. During simulation, source equations for signals can be requested for display either through direct input of a signal name or through graphical interface with the simulation display. Signal equations can in this manner be traced back through several levels, which conveniently provides important information during the observation and modification of signal values. Memory areas may be associated with a processor and loaded with data to be executed by that processor during the simulation. The data can be displayed in both numerical and assembly code mnemonic form, and may also be modified by entering numbers or assembly instructions. The simulated processor execution may thus be interactively modified during the simulation. Additionally, the ability to step the simulation backwards through time, and then forward again, is provided.
摘要:
A computer system simulator concurrently models both processor operation and signal logic behavior and provides a high degree of user interaction and flexibility in the observation and control of signal values and memory contents during the execution of a simulation. During simulation, source equations for signals can be requested for display either through direct input of a signal name or through graphical interface with the simulation display. Signal equations can in this manner be traced back through several levels, which conveniently provides important information during the observation and modification of signal values. Memory areas may be associated with a processor and loaded with data to be executed by that processor during the simulation. The data can be displayed in both numerical and assembly code mnemonic form, and may also be modified by entering numbers or assembly instructions. The simulated processor execution may thus be interactively modified during the simulation. Additionally, the ability to step the simulation backwards through time, and then forward again, is provided.
摘要:
A parallel processing system is provided with a plurality of processors and a plurality of memories, and bus units with arbitration coupling the processors and memories. A bus unit provides a pathway between one processor and the bus unit's respective memory. Each bus unit arbitrates multiple simultaneous access requests for its respective memory and communicates its decisions to other bus units so that a memory access requiring multiple memories will only occur if all those memories are available. The coupling of processors to memories can change, dynamically, each bus cycle without the need for setup before the bus cycle either by pipelining or having unused bus cycles. In a specific embodiment, the memory access information is provided on high order address lines, where the processor logically accesses different memory address spaces to make different accesses, thereby sharing memory with other processors.
摘要:
The present invention provides for a computer system having a plurality of parallel processor units. The processor units are connected in common to a signal line with each processor capable of setting a first signal level on the line and monitoring the line in response to instructions to the processor. This allows each processor unit to be notified of the completion of a parallel operation by other participating processor units upon a second signal level on the signal line. More than one signal lines may be connected between the parallel processor units to provide synchronization of different parallel operations between different processor units.
摘要:
A computer system having a plurality of parallel processor units with each processor unit having an output bus of n bits and an associated mask register is provided. The computer system comprises a bus unit, coupled to the output bus of each processor unit and each associated mask register, for masking the output bus bits with bits in the mask register of each processor unit and logically combining the resulting masked bits from each processor unit into an output bus of n bits in one computer operation.
摘要:
A computer system having a plurality of parallel processor units with each processor unit having an output bus of n bits and an associated mask register is provided. The computer system comprises a bus unit, coupled to the output bus of each processor unit and each associated mask register, for masking the output bus bits with bits in the mask register of each processor unit and logically combining the resulting masked bits from each processor unit into an output bus of n bits in one computer operation.
摘要:
A simulation of an electronics system which performs a set of operations of interest. A simulated supervisory circuit detects a state in which all the operations have been completed, and also determines the amount of time until the occurrence of the next relevant event. Simulation time is then advanced by that amount of time. This enables simulation time corresponding to an inactive system to be eliminated.
摘要:
A statistically fast, high performance computer memory system including a system memory for storing code and non-code data accessible by at least two bus masters, a bus connecting the memory with the bus masters, and a plurality of caches connected to the bus. An internal cache holds data selected solely on the basis of memory accesses by the host processor, a pre-fetch cache pre-fetches code from the memory, and a write buffer cache connected to the bus for buffering data written to the memory.
摘要:
A method and system for transferring data elements from a computer main memory to a cache memory. The main and cache memories are accessible by a host processor and other bus masters connected thereto by a bus. Code data elements to be read by the host processor are predicted. The predicted code data elements are then transferred from the main memory to cache memory without delaying memory access requests for data from the other bus masters.