Wang Laboratories, Inc.

James B MacDonald

Stephen W Olson

Edward D Mann

Robert R Trottier

James W Petersen, Jr

David L Whipple

Kevin T Mahoney

Dennis J Kayser

Peter A Morrison

Robert C Nash

Tom R Kibler

Bhikoo J Patel

Andrew N Fu

Richard W Lones

John M Martins

Disclosed are methods and apparatus for interfacing a central processor (12) (CP) and an IO controller (30) (IOC) to a main memory (40). A CP and an IO write buffer each include a pair of memory input data registers, located in a pair of Memory Data Unit (MDU) integrated circuits (38a, 38b), and also two memory address registers, a previous memory address register, and an address comparator, located in a Memory Address Unit (MAU) (36). These registers, in conjunction with associated control logic, are used to buffer CP and IO write addresses and data to the main memory. If both address registers have a pending write, the last loaded address register is checked for a match against the current write address using the previous address register and the comparator. A match results in the combination of the previous write data and the current write data into one pending write, using write merge circuitry within the MDUs. IO read operations are performed by first loading the MDU read data registers with read data from memory locations specified by a quad-word aligned address in the first IO address register, followed by an incremented quad-word aligned address in the second IO address register. The transfer is then begun and all incoming IO read addresses are checked for a current quad-word compare. If an incoming quad-word aligned IO read address is not equal to the content of the IO previous address register, a memory read request is generated using the incremented address, and the MDU read data registers are advanced. A feature of this invention is that no specific addresses are used, and a knowledge of a transfer width (byte, word, etc.) is not required to determine memory operation types.

Apparatus and methods for reducing numbers of read-modify-write cycles

A circuit (10) for generating an output signal having a frequency that is a multiple of an input clock signal (CLKIN). The circuit includes a delay circuit (12) having an input port and a plurality of output ports (A,B,C). The input port is coupled during use to the input clock signal. Individual ones of the plurality of output ports output a signal that is delayed with respect to the input clock signal and also with respect to others of the plurality of output ports. The circuit further includes a logic network (20) having a first input for coupling to the input clock signal and a plurality of second inputs for coupling to the plurality of output ports. The logic network operates to logically combine signals emanating from the plurality of output ports with the input clock signal, and has an output port (OUTPUT) for outputting a signal having a frequency that is multiple of a frequency of the input clock signal. The signal that is output from the output port of the logic network has a 50% duty cycle regardless of the duty cycle of the input clock signal.

Clock frequency multiplying and squaring circuit and method

A method, and circuitry that operates in accordance with the method, for generating an entry for a translation buffer in a data processor that employs virtual memory addressing. The method includes the first steps of storing a Faulted Virtual Address in a first register (96) and a Zone Table Address (ZTA) in a second register (94). In response to the execution of a micro-instruction, a next step forms an address in memory of a Zone Table Entry (ZTE) by selectively combining a first portion of the content of the first register with the content of the second register, while simultaneously testing the ZTA for physical address mapping. In response to an execution of a next micro-instruction, a next step accesses the ZTE with the formed address, and forms an address in memory of a Segment Table Entry (STE) by selectively combining a second portion the content of the first register with a content of the ZTE, while simultaneously testing the ZTE for a Zone fault. In response to an execution of a next micro-instruction, a next step accesses the STE with the formed address, and forms an address in memory of a Page Table Entry (PTE) by selectively combining a third portion of the content of the first register with a content of the STE, while simultaneously testing the STE for a Zone fault. In response to an execution of a next micro-instruction, a next step accesses the PTE with the formed address and selectively combines the content of the STE with the content of the PTE and outputs the combination as the translation buffer entry, while simultaneously testing the PTE for a Page fault.

Virtual address translation hardware assist circuit and method

An A output and a B output of a register file 16 are each provided to an associated multiplexer (18,20). Each multiplexer has as a further input a bus (CB00:31) that conveys a result from an ALU 22 via an ALU shifter 28. Outputs of the multiplexers are provided to corresponding A or B inputs of the ALU. Each multiplexer is controlled by an associated register file address comparator (24,26). The address comparators each have as an input corresponding register file A and B update and access addresses. The address comparators compare their associated register file update and access addresses to determine if the register file register selected for access is equal to the register file register selected for update. If these two addresses are found to be equal it is indicated that the result of an ALU operation during an instruction cycle N is to be used as an operand for an ALU operation during a cycle N+1. When this condition is detected the output of the associated address comparator enables the corresponding multiplexer select input to gate the ALU result directly to the corresponding input of the ALU, thereby effectively bypassing the register file.

Improved CPU pipeline having register file bypass and working register

An I/O structure for use in a digital data processing system of the type in which system components including a processor and a system memory are connected by a system bus. The I/O structure includes a system bus interface connected to the system bus, a synchronous satellite processing unit (SPU) bus connected to the system bus interface, one or more satellite processing units (SPUs) connected to the SPU bus, and peripheral devices attached to the satellite processing units. Each SPU has three main components: control logic including a microprocessor for controlling the SPU, a device adapter specific to the peripheral device for controlling the peripheral device and transferring data between the peripheral device and the SPU, and an interface unit connected to the control logic and the device adapter for providing I/O communications to the SPU bus and responding to I/O communications on the SPU bus. The I/O communications fall into two classes: communications to SPUs and communications to system components. The communications to SPUs all require a single SPU bus cycle; the communications to system components require one or more cycles. The system bus interface translates communications to system components into communications on the system bus and translates communications on the system bus intended for a SPU into communications to SPUs. The SPU bus includes first lines for carrying an I/O command and an identifier for an SPU involved in the communication and second lines for carrying the contents of the communication. In multicycle communications, the I/O command and identifier remain on the first lines for all cycles, but the information on the second lines varies from cycle to cycle.

I/O structure for information processing system

Apparatus and method for reconfiguring a memory in a data processing system to increase the rate of information transfer between system memory and processor. The system memory is comprised of a plurality M of memory banks, each having a separate data output path. In a first configuration a memory controller addresses the memory banks sequentially to read from one address location at a time. The memory is reconfigured by an address translator providing addresses addressing M banks in parallel, so that M locations are read in each read operation, and a bus reconfiguration multiplexer which reconfigures the bank output busses in parallel and selects one or more bank output busses as the memory output to the system processor. In a further embodiment, a cache is connected in parallel with the parallel bank outputs for storing information appearing upon the non-selected bank output busses paths for subsequent transfer to the memory output in a subsequent reading from memory of previously read but non-selected information.

Growing community of inventors

Dracut, MA, United States of America

James B MacDonald