A response priority circuit arrangement for determining the priority among simultaneous responses from different access-time levels L3 and L2 in a storage hierarchy to maintain nearly the same order among the simultaneous responses as the order of their storage access requests. The storage requests were put into an indexed slot in a hardware queue. The index of the assigned queue slot is sent to a basic storage module (BSM) part of the hierarchy which is selected by the storage address supplied by the processor making the storage request. The selected BSM will have the requested data either in its main memory part (L3) or in its high-speed buffer part (L2). The priority circuit arrangement has a separate group of AND gates for each hierarchy level L3 and L2. The groups are interlocked by a circuit which disables the L2 group if any AND gate is enabled in the L3 group. Only one AND gate can be enabled in both L3 and L2 groups. When simultaneous responses are signalled from the L3 and L2 levels, the L3 gate is given priority since the L3 response resulted from an earlier request than the L2 response. The output of the enabled AND gate indicates the index of the slot in the queue which contains the request matching the response given priority. A bus connection can then be made using that slot's information for the data transfer between the selected BSM and the requesting processor.
A computer system which utilizes processor boards including a first level cache system integrated with the microprocessor, a second level external cache system and a third level external cache system. The second level cache system is a conventional, high speed, SRAM-based, writeback cache system. The third level cache system is a large, writethrough cache system developed using conventional DRAMs as used in the main memory subsystem of the computer system. The three cache systems are arranged between the CPU and the host bus in a serial fashion. Because of the large size of the third level cache, a high hit rate is developed so that operations are not executed on the host bus but are completed locally on the processor board, reducing the use of the host bus by an individual processor board. This allows additional processor boards to be installed in the computer system without saturating the host bus. The third level cache system is organized as a writethrough cache. However, the shared or exclusive status of any cached data is also stored. If the second level cache performs a write allocate cycle and the data is exclusive in the third level cache, the data is provided directly from the third level cache, without requiring an access to main memory, reducing the use of the host bus.
A shared memory computer method and apparatus having a plurality of sources, a memory manager, and memory units in which the memory locations of data items are randomly distributed. The memory manager includes a translation module for locating data items in the memory units and a temporary storage buffer for storing at least a portion of messages between sources and the memory units with respect to data items.
A grouping device comprises a register table and a grouping unit the register table having m registers corresponding to m groups, each register including an n-bits data storing portion corresponding to the n input signals, for registering relationships between the n input signals and the m groups, the grouping unit receiving grouping signals output from the register table and the n input signals, for selecting one group from the m groups for each input signal and grouping each input signal into the selected group in accordance with the register table. Therefore, the register access time is shortened and the confirmation of the contents of the register by the CPU is made easier.
An input/output system includes a plurality of modules and a system interface unit having a plurality of ports, each of which connects to a different one of the modules. The plurality of modules includes at least one processor and one memory module. The system interface unit includes a timer unit and a priority network for processing processor interrupt requests on a priority basis. The priority network connects to a register for storing coded priority level signals to be assigned to the different types of interrupt requests. The register is conditioned to store a low priority level for timer interrupts. The timer unit includes a preset register, an interval counter and a rollover counter. At the completion of each time interval, the interval counter is loaded automatically from the preset register and counting is continued. Simultaneously, the interval counter conditions the rollover counter to store a count registering the total number of completed intervals counted. The processor responds to a timer interrupt request by a special command which can only selectively clear the rollover counter to ZEROS when the interval counter is not in the process of completing another interval.
A multilevel cache buffer for a multiprocessor system in which each processor has a level one cache storage unit which interfaces with a level two cache unit and main storage unit shared by all processors. The multiprocessors share the level two cache according to a priority algorithm. When data in the level two cache is updated, corresponding data in level one caches is invalidated until it is updated.
