Amdahl Corporation

Other

Allan J Zmyslowski

Robert M Maier

Carolee V Newcomb

Quang H Nguyen

Chris Norrie

Arno S Krakauer

Richard J Tobias

John C Andoh

Arun Shah

Joseph A Petolino

Stephen J Rawlinson

Jeffrey A Thomas

Michael DeMar Taylor

Christopher I Norrie

Edward G Ryba

Peter K Yu

Michael J Begley

Gary A Browning

Luis Ancajas

Carolee N Schober

Jon K Lexau

Pat Y Hom

Robert A Shaw

A data integrity system comprising a plurality of units connected together for the transfer of data between the units. Each said unit comprises data means for receiving data from one or more other units and/or transmitting data to one or more other units, first means, if the unit transmits data to other units, for generating a PARITY OUT signal indicating the parity of all data transmitted to the other units and second means, if the unit is to receive data from other units, for generating a PARITY IN signal indicating the parity of all data being received by the unit from the other units. A third means, located in one of said plurality of units, receives all the PARITY OUT signals and all the PARITY IN signals generated by the plurality of units and detects from all the received PARITY IN signals and all the PARITY OUT signals whether an error had occurred during the transfer of data between the units.

Parity checking system with reduced usage of I/O pins

A processing system comprising a first means for generating first signals indicating when the next instruction can begin processing where eight or less bytes are processed by the MOVE, PACK or UNPACK instruction, a second means for generating second signals if an overlap condition exists for the MOVE, PACK or UNPACK instruction being processed, and where the first means generates the first signals prior to the second means generating the second signals and independent of whether the second means generates the second signals.

System for processing eight bytes or less by the move, pack and unpack

A random access memory (RAM) complex that can concurrently read and write to different addresses. The memory complex includes two RAMs, each having an address selector, includes a data out multiplexer for selecting outputs from one of the RAM's. A tag array stores an array of tag, one for each address in the RAM's. The tag marks which one of the two RAM's has the valid data for the corresponding read address. During a concurrent read and write cycle, the tag selects the read address for one RAM, selects the write address for the other RAM and a staged copy of the tag controls the multiplexer to select data from the correct RAM for the data out.

Memory having concurrent read and writing from different addresses

An error recovery system used in a pipeline architecture type computer system for recovering from an error in a control word for an instruction without interrupting the sequence of processing control words by the computer system. The computer system processes instructions in a sequence of overlapping FLOWs where each FLOW is comprised of a sequence of cycles. An instruction control word is processed in each cycle of each FLOW. The error recovery system comprises a first storage for storing, for a given cycle of a FLOW, all the control words for all the instructions, a second storage for storing a control word read from the first storage and an error recovery logic for detecting an error in the control word read from the first storage and stored in the second storage and for correcting the error in the control word in the first and second storage.

Error recovery system for recovering errors that occur in control store

A timer unit that permits individual timer registers to be taken offline from the timer complex. A single register is taken offline instead of checkstopping the entire computer system due to a damaged timer, for example, thereby reducing system outages and thus providing increased availability of the system.

Method for reconfiguring individual timer registers offline

A computer system having a number of circuits and error detection circuitry for detecting errors in the circuits and providing error signals. Error recovery is provided by detecting an error with checking logic, turning off system clocks and invoking error recovery software, executing recovery software to recover from the error and set an action latch, turning on system clocks, taking hardware recovery action in response to the action latch, resetting the action latch and continue processing. Action latches are used extensively throughout the computer system.

Invoking hardware recovery actions via action latches

In a data processing apparatus, having a user domain with domain storage space and an emulation domain with emulation storage space, emulates an instruction in the user domain by calling a program of instructions in the emulation domain. An instruction register connected to receive the sequence of instructions is partitioned into a plurality of fields. At least one field of the instruction register identifies a location for address information for an operand used in execution of the instruction. The instruction is decoded to generate a control code. The control code includes a branch signal to call the emulation program and a domain access control signal to indicate whether the emulation program requires access to the user domain storage base for execution. During execution of the emulation routine, domain access is implied by the location identified for the address information--a set of locations having been preselected as address registers for implied domain addressing.

Multiple domain emulation system with separate domain facilities which

A central processor architecture implementating a deterministic, digit based, subterm computation and selective subterm combination early condition code analysis mechanism to provide for the early determination of the condition code that will be returned upon normal execution of a condition code setting instruction is described. The early condition code is returned sufficiently early to be combined with a branch mask to substantially reduce if not eliminate the need to interlock an immediately following conditional branch instruction. A wide variety of condition code setting instructions are handled by the deterministic condition code analysis mechanism of the present invention by implementing the mechanism to determine condition codes by the generation of digit subterms of the operand data accompanying condition code setting instruction and then combining the digit subterms in a predetermined manner selected based on the specific condition code setting instruction being executed. Branch decision analysis is performed by inclusion of the branch decision mask as a subterm selectively combined with the condition code subterms.

Central processor architecture implementing deterministic early

An apparatus and method for use in improving cache storage unit utilization during an interlock of an instruction pipeline generates a control signal during one cycle of the interlock if the interlocked instruction may require storage unit management work. In response to the control signal, selector control logic in the storage unit generates a priority signal indicating the interlocked instruction for selection by the storage unit for processing. In response to the control signal and the priority signal, the cache management logic is used during the interlock on the interlocked instruction to prepare for supplying needed data when the interlock is released.

Apparatus and method for improving cache access throughput in pipelined

An interlock of an instruction processing pipeline in a data processing system responsive to the validity of the pipeline stages within the instruction unit pipeline under microprogram control, is provided. Thus, a microprogram can provide for the release of a particular pipeline stage based on a selected characteristic of the valid signals generated by other stages of the pipeline. An interlock control signal is generated by a decode of a field in a microinstruction stored in a control store RAM or through hardwired decoding.

Microprogrammable pipeline interlocks based on the validity of pipeline

The present invention provides for use in a data processor a method for mapping a respective machine language instruction stored by cache storage unit to a respective microprogrammed algorithm stored in control storage unit means, wherein the respective machine language instruction includes an opcode field with a prescribed value and at least one nonopcode field with one of a plurality of values, the method, comprising the steps of in the course of one data processor clock cycle, providing the respective machine language instruction to a decoder for converting the prescribed opcode field and the at least one nonopcode field of the respective machine language instruction into a respective combination of decoded signals which corresponds to the prescribed opcode field value and that at least one nonopcode field value of the respective machine language instruction; and providing the respective combination of decoded signals to combinational logic for converting the respective combination of decode signals into a respective address signal which points to a microprogrammed algorithm stored by the control storage unit means.

Growing community of inventors

Sunnyvale, CA, United States of America

Allan J Zmyslowski